diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,65844 @@ +{ + "best_global_step": 38400, + "best_metric": 0.08336079865694046, + "best_model_checkpoint": "saves/prompt-tuning/gemma-3-1b-it/train_qqp_1744902593/checkpoint-38400", + "epoch": 1.9544377397210075, + "eval_steps": 200, + "global_step": 40000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.00024430166369432974, + "grad_norm": 4.755948066711426, + "learning_rate": 0.29999999259779675, + "loss": 8.9878, + "num_input_tokens_seen": 6432, + "step": 5 + }, + { + "epoch": 0.0004886033273886595, + "grad_norm": 4.295941352844238, + "learning_rate": 0.29999996252634736, + "loss": 6.6033, + "num_input_tokens_seen": 13184, + "step": 10 + }, + { + "epoch": 0.0007329049910829893, + "grad_norm": 2.3181283473968506, + "learning_rate": 0.2999999093230187, + "loss": 3.7242, + "num_input_tokens_seen": 19808, + "step": 15 + }, + { + "epoch": 0.000977206654777319, + "grad_norm": 0.6410034894943237, + "learning_rate": 0.299999832987819, + "loss": 2.0963, + "num_input_tokens_seen": 26432, + "step": 20 + }, + { + "epoch": 0.0012215083184716488, + "grad_norm": 0.311041921377182, + "learning_rate": 0.29999973352076004, + "loss": 0.9446, + "num_input_tokens_seen": 32896, + "step": 25 + }, + { + "epoch": 0.0014658099821659785, + "grad_norm": 0.10064440220594406, + "learning_rate": 0.2999996109218572, + "loss": 0.3982, + "num_input_tokens_seen": 39072, + "step": 30 + }, + { + "epoch": 0.0017101116458603082, + "grad_norm": 2.0101823806762695, + "learning_rate": 0.2999994651911293, + "loss": 0.4405, + "num_input_tokens_seen": 45312, + "step": 35 + }, + { + "epoch": 0.001954413309554638, + "grad_norm": 0.32684221863746643, + "learning_rate": 0.2999992963285989, + "loss": 0.36, + "num_input_tokens_seen": 51680, + "step": 40 + }, + { + "epoch": 0.002198714973248968, + "grad_norm": 0.16424602270126343, + "learning_rate": 0.29999910433429194, + "loss": 0.2332, + "num_input_tokens_seen": 57952, + "step": 45 + }, + { + "epoch": 0.0024430166369432977, + "grad_norm": 0.09016671031713486, + "learning_rate": 0.29999888920823814, + "loss": 0.2562, + "num_input_tokens_seen": 64256, + "step": 50 + }, + { + "epoch": 0.0026873183006376274, + "grad_norm": 0.19881433248519897, + "learning_rate": 0.29999865095047057, + "loss": 0.2613, + "num_input_tokens_seen": 71264, + "step": 55 + }, + { + "epoch": 0.002931619964331957, + "grad_norm": 0.15641289949417114, + "learning_rate": 0.29999838956102604, + "loss": 0.2828, + "num_input_tokens_seen": 77952, + "step": 60 + }, + { + "epoch": 0.0031759216280262867, + "grad_norm": 0.12226133048534393, + "learning_rate": 0.29999810503994484, + "loss": 0.2034, + "num_input_tokens_seen": 84416, + "step": 65 + }, + { + "epoch": 0.0034202232917206164, + "grad_norm": 0.03750175982713699, + "learning_rate": 0.29999779738727084, + "loss": 0.234, + "num_input_tokens_seen": 90880, + "step": 70 + }, + { + "epoch": 0.0036645249554149465, + "grad_norm": 0.04839527979493141, + "learning_rate": 0.29999746660305154, + "loss": 0.1966, + "num_input_tokens_seen": 97568, + "step": 75 + }, + { + "epoch": 0.003908826619109276, + "grad_norm": 0.04795560985803604, + "learning_rate": 0.2999971126873379, + "loss": 0.2096, + "num_input_tokens_seen": 103968, + "step": 80 + }, + { + "epoch": 0.004153128282803606, + "grad_norm": 0.030905520543456078, + "learning_rate": 0.2999967356401845, + "loss": 0.1771, + "num_input_tokens_seen": 110400, + "step": 85 + }, + { + "epoch": 0.004397429946497936, + "grad_norm": 0.030110904946923256, + "learning_rate": 0.29999633546164944, + "loss": 0.1922, + "num_input_tokens_seen": 116768, + "step": 90 + }, + { + "epoch": 0.004641731610192265, + "grad_norm": 0.06372471898794174, + "learning_rate": 0.29999591215179444, + "loss": 0.17, + "num_input_tokens_seen": 123232, + "step": 95 + }, + { + "epoch": 0.004886033273886595, + "grad_norm": 0.047347236424684525, + "learning_rate": 0.2999954657106849, + "loss": 0.1701, + "num_input_tokens_seen": 129536, + "step": 100 + }, + { + "epoch": 0.005130334937580925, + "grad_norm": 0.03338063508272171, + "learning_rate": 0.2999949961383896, + "loss": 0.1572, + "num_input_tokens_seen": 136448, + "step": 105 + }, + { + "epoch": 0.005374636601275255, + "grad_norm": 0.030674610286951065, + "learning_rate": 0.2999945034349809, + "loss": 0.1772, + "num_input_tokens_seen": 143168, + "step": 110 + }, + { + "epoch": 0.005618938264969585, + "grad_norm": 0.017095131799578667, + "learning_rate": 0.2999939876005348, + "loss": 0.1649, + "num_input_tokens_seen": 149696, + "step": 115 + }, + { + "epoch": 0.005863239928663914, + "grad_norm": 0.023525789380073547, + "learning_rate": 0.29999344863513094, + "loss": 0.156, + "num_input_tokens_seen": 156832, + "step": 120 + }, + { + "epoch": 0.006107541592358244, + "grad_norm": 0.021489722654223442, + "learning_rate": 0.2999928865388523, + "loss": 0.1833, + "num_input_tokens_seen": 163456, + "step": 125 + }, + { + "epoch": 0.0063518432560525735, + "grad_norm": 0.01983758993446827, + "learning_rate": 0.29999230131178567, + "loss": 0.1657, + "num_input_tokens_seen": 169920, + "step": 130 + }, + { + "epoch": 0.006596144919746904, + "grad_norm": 0.024278059601783752, + "learning_rate": 0.2999916929540212, + "loss": 0.1588, + "num_input_tokens_seen": 176480, + "step": 135 + }, + { + "epoch": 0.006840446583441233, + "grad_norm": 0.016512328758835793, + "learning_rate": 0.29999106146565285, + "loss": 0.1754, + "num_input_tokens_seen": 182464, + "step": 140 + }, + { + "epoch": 0.007084748247135563, + "grad_norm": 0.014139589853584766, + "learning_rate": 0.29999040684677786, + "loss": 0.1334, + "num_input_tokens_seen": 188800, + "step": 145 + }, + { + "epoch": 0.007329049910829893, + "grad_norm": 0.01790541596710682, + "learning_rate": 0.2999897290974972, + "loss": 0.1418, + "num_input_tokens_seen": 195424, + "step": 150 + }, + { + "epoch": 0.007573351574524222, + "grad_norm": 0.032509054988622665, + "learning_rate": 0.2999890282179155, + "loss": 0.1626, + "num_input_tokens_seen": 201952, + "step": 155 + }, + { + "epoch": 0.007817653238218552, + "grad_norm": 0.026784060522913933, + "learning_rate": 0.29998830420814077, + "loss": 0.1572, + "num_input_tokens_seen": 209408, + "step": 160 + }, + { + "epoch": 0.008061954901912883, + "grad_norm": 0.019511893391609192, + "learning_rate": 0.2999875570682846, + "loss": 0.1635, + "num_input_tokens_seen": 215808, + "step": 165 + }, + { + "epoch": 0.008306256565607212, + "grad_norm": 0.012516159564256668, + "learning_rate": 0.2999867867984623, + "loss": 0.1317, + "num_input_tokens_seen": 222496, + "step": 170 + }, + { + "epoch": 0.008550558229301541, + "grad_norm": 0.01747675985097885, + "learning_rate": 0.29998599339879267, + "loss": 0.1559, + "num_input_tokens_seen": 229152, + "step": 175 + }, + { + "epoch": 0.008794859892995872, + "grad_norm": 0.018156783655285835, + "learning_rate": 0.29998517686939796, + "loss": 0.1467, + "num_input_tokens_seen": 235712, + "step": 180 + }, + { + "epoch": 0.009039161556690201, + "grad_norm": 0.031365249305963516, + "learning_rate": 0.29998433721040413, + "loss": 0.1557, + "num_input_tokens_seen": 242048, + "step": 185 + }, + { + "epoch": 0.00928346322038453, + "grad_norm": 0.030253317207098007, + "learning_rate": 0.29998347442194073, + "loss": 0.179, + "num_input_tokens_seen": 248352, + "step": 190 + }, + { + "epoch": 0.00952776488407886, + "grad_norm": 0.009556518867611885, + "learning_rate": 0.2999825885041407, + "loss": 0.1618, + "num_input_tokens_seen": 254688, + "step": 195 + }, + { + "epoch": 0.00977206654777319, + "grad_norm": 0.030872635543346405, + "learning_rate": 0.29998167945714077, + "loss": 0.1845, + "num_input_tokens_seen": 260832, + "step": 200 + }, + { + "epoch": 0.00977206654777319, + "eval_loss": 0.18648895621299744, + "eval_runtime": 401.3836, + "eval_samples_per_second": 90.649, + "eval_steps_per_second": 22.664, + "num_input_tokens_seen": 260832, + "step": 200 + }, + { + "epoch": 0.01001636821146752, + "grad_norm": 0.05658264830708504, + "learning_rate": 0.2999807472810811, + "loss": 0.2019, + "num_input_tokens_seen": 267232, + "step": 205 + }, + { + "epoch": 0.01026066987516185, + "grad_norm": 0.02793041430413723, + "learning_rate": 0.29997979197610536, + "loss": 0.1971, + "num_input_tokens_seen": 273472, + "step": 210 + }, + { + "epoch": 0.01050497153885618, + "grad_norm": 0.030114196240901947, + "learning_rate": 0.299978813542361, + "loss": 0.1805, + "num_input_tokens_seen": 280096, + "step": 215 + }, + { + "epoch": 0.01074927320255051, + "grad_norm": 0.023829180747270584, + "learning_rate": 0.2999778119799988, + "loss": 0.22, + "num_input_tokens_seen": 286912, + "step": 220 + }, + { + "epoch": 0.010993574866244839, + "grad_norm": 0.015243886038661003, + "learning_rate": 0.29997678728917326, + "loss": 0.1597, + "num_input_tokens_seen": 293216, + "step": 225 + }, + { + "epoch": 0.01123787652993917, + "grad_norm": 0.033879246562719345, + "learning_rate": 0.2999757394700424, + "loss": 0.1513, + "num_input_tokens_seen": 299744, + "step": 230 + }, + { + "epoch": 0.011482178193633499, + "grad_norm": 0.01922355592250824, + "learning_rate": 0.29997466852276783, + "loss": 0.1669, + "num_input_tokens_seen": 306240, + "step": 235 + }, + { + "epoch": 0.011726479857327828, + "grad_norm": 0.018462467938661575, + "learning_rate": 0.29997357444751466, + "loss": 0.1688, + "num_input_tokens_seen": 312800, + "step": 240 + }, + { + "epoch": 0.011970781521022157, + "grad_norm": 0.01671449840068817, + "learning_rate": 0.2999724572444516, + "loss": 0.1652, + "num_input_tokens_seen": 319200, + "step": 245 + }, + { + "epoch": 0.012215083184716488, + "grad_norm": 0.016798265278339386, + "learning_rate": 0.29997131691375095, + "loss": 0.1608, + "num_input_tokens_seen": 325856, + "step": 250 + }, + { + "epoch": 0.012459384848410818, + "grad_norm": 0.013317232020199299, + "learning_rate": 0.2999701534555886, + "loss": 0.1341, + "num_input_tokens_seen": 332064, + "step": 255 + }, + { + "epoch": 0.012703686512105147, + "grad_norm": 0.014112112112343311, + "learning_rate": 0.2999689668701439, + "loss": 0.1568, + "num_input_tokens_seen": 338400, + "step": 260 + }, + { + "epoch": 0.012947988175799478, + "grad_norm": 0.00875961221754551, + "learning_rate": 0.29996775715759993, + "loss": 0.1638, + "num_input_tokens_seen": 344864, + "step": 265 + }, + { + "epoch": 0.013192289839493807, + "grad_norm": 0.01935550384223461, + "learning_rate": 0.2999665243181432, + "loss": 0.1576, + "num_input_tokens_seen": 351232, + "step": 270 + }, + { + "epoch": 0.013436591503188136, + "grad_norm": 0.013917384669184685, + "learning_rate": 0.2999652683519638, + "loss": 0.1706, + "num_input_tokens_seen": 358144, + "step": 275 + }, + { + "epoch": 0.013680893166882466, + "grad_norm": 0.011072104796767235, + "learning_rate": 0.29996398925925544, + "loss": 0.1683, + "num_input_tokens_seen": 364832, + "step": 280 + }, + { + "epoch": 0.013925194830576797, + "grad_norm": 0.009779469110071659, + "learning_rate": 0.2999626870402154, + "loss": 0.153, + "num_input_tokens_seen": 371488, + "step": 285 + }, + { + "epoch": 0.014169496494271126, + "grad_norm": 0.01639954373240471, + "learning_rate": 0.29996136169504445, + "loss": 0.1724, + "num_input_tokens_seen": 377984, + "step": 290 + }, + { + "epoch": 0.014413798157965455, + "grad_norm": 0.023673513904213905, + "learning_rate": 0.29996001322394694, + "loss": 0.1504, + "num_input_tokens_seen": 384256, + "step": 295 + }, + { + "epoch": 0.014658099821659786, + "grad_norm": 0.0189872607588768, + "learning_rate": 0.29995864162713093, + "loss": 0.2095, + "num_input_tokens_seen": 390400, + "step": 300 + }, + { + "epoch": 0.014902401485354115, + "grad_norm": 0.008692477829754353, + "learning_rate": 0.2999572469048079, + "loss": 0.161, + "num_input_tokens_seen": 396576, + "step": 305 + }, + { + "epoch": 0.015146703149048445, + "grad_norm": 0.0352720208466053, + "learning_rate": 0.29995582905719287, + "loss": 0.1802, + "num_input_tokens_seen": 403200, + "step": 310 + }, + { + "epoch": 0.015391004812742774, + "grad_norm": 0.016630228608846664, + "learning_rate": 0.2999543880845046, + "loss": 0.1652, + "num_input_tokens_seen": 409536, + "step": 315 + }, + { + "epoch": 0.015635306476437103, + "grad_norm": 0.04912297800183296, + "learning_rate": 0.2999529239869652, + "loss": 0.1632, + "num_input_tokens_seen": 416128, + "step": 320 + }, + { + "epoch": 0.015879608140131434, + "grad_norm": 0.020790671929717064, + "learning_rate": 0.2999514367648005, + "loss": 0.1719, + "num_input_tokens_seen": 422336, + "step": 325 + }, + { + "epoch": 0.016123909803825765, + "grad_norm": 0.02688586711883545, + "learning_rate": 0.29994992641823987, + "loss": 0.1691, + "num_input_tokens_seen": 429152, + "step": 330 + }, + { + "epoch": 0.016368211467520093, + "grad_norm": 0.010249392129480839, + "learning_rate": 0.29994839294751613, + "loss": 0.1681, + "num_input_tokens_seen": 435488, + "step": 335 + }, + { + "epoch": 0.016612513131214424, + "grad_norm": 0.01339343935251236, + "learning_rate": 0.29994683635286584, + "loss": 0.1555, + "num_input_tokens_seen": 441952, + "step": 340 + }, + { + "epoch": 0.016856814794908755, + "grad_norm": 0.0229802168905735, + "learning_rate": 0.2999452566345291, + "loss": 0.1895, + "num_input_tokens_seen": 448672, + "step": 345 + }, + { + "epoch": 0.017101116458603082, + "grad_norm": 0.008547558449208736, + "learning_rate": 0.2999436537927494, + "loss": 0.1577, + "num_input_tokens_seen": 454720, + "step": 350 + }, + { + "epoch": 0.017345418122297413, + "grad_norm": 0.011749034747481346, + "learning_rate": 0.299942027827774, + "loss": 0.1537, + "num_input_tokens_seen": 461216, + "step": 355 + }, + { + "epoch": 0.017589719785991744, + "grad_norm": 0.008608849719166756, + "learning_rate": 0.29994037873985363, + "loss": 0.1495, + "num_input_tokens_seen": 467584, + "step": 360 + }, + { + "epoch": 0.01783402144968607, + "grad_norm": 0.021838665008544922, + "learning_rate": 0.29993870652924254, + "loss": 0.1587, + "num_input_tokens_seen": 473696, + "step": 365 + }, + { + "epoch": 0.018078323113380403, + "grad_norm": 0.010592716746032238, + "learning_rate": 0.29993701119619876, + "loss": 0.1707, + "num_input_tokens_seen": 480288, + "step": 370 + }, + { + "epoch": 0.018322624777074734, + "grad_norm": 0.011896399781107903, + "learning_rate": 0.2999352927409835, + "loss": 0.1548, + "num_input_tokens_seen": 486720, + "step": 375 + }, + { + "epoch": 0.01856692644076906, + "grad_norm": 0.009385009296238422, + "learning_rate": 0.29993355116386194, + "loss": 0.1491, + "num_input_tokens_seen": 493120, + "step": 380 + }, + { + "epoch": 0.018811228104463392, + "grad_norm": 0.017876043915748596, + "learning_rate": 0.29993178646510266, + "loss": 0.1658, + "num_input_tokens_seen": 499456, + "step": 385 + }, + { + "epoch": 0.01905552976815772, + "grad_norm": 0.007711249869316816, + "learning_rate": 0.2999299986449777, + "loss": 0.1663, + "num_input_tokens_seen": 505792, + "step": 390 + }, + { + "epoch": 0.01929983143185205, + "grad_norm": 0.009576726704835892, + "learning_rate": 0.29992818770376284, + "loss": 0.1748, + "num_input_tokens_seen": 512064, + "step": 395 + }, + { + "epoch": 0.01954413309554638, + "grad_norm": 0.0064466060139238834, + "learning_rate": 0.29992635364173725, + "loss": 0.1357, + "num_input_tokens_seen": 518880, + "step": 400 + }, + { + "epoch": 0.01954413309554638, + "eval_loss": 0.15296052396297455, + "eval_runtime": 402.8217, + "eval_samples_per_second": 90.325, + "eval_steps_per_second": 22.583, + "num_input_tokens_seen": 518880, + "step": 400 + }, + { + "epoch": 0.01978843475924071, + "grad_norm": 0.01684892550110817, + "learning_rate": 0.2999244964591839, + "loss": 0.1414, + "num_input_tokens_seen": 526240, + "step": 405 + }, + { + "epoch": 0.02003273642293504, + "grad_norm": 0.007950413972139359, + "learning_rate": 0.2999226161563891, + "loss": 0.1737, + "num_input_tokens_seen": 532768, + "step": 410 + }, + { + "epoch": 0.02027703808662937, + "grad_norm": 0.012899374589323997, + "learning_rate": 0.2999207127336429, + "loss": 0.1708, + "num_input_tokens_seen": 539296, + "step": 415 + }, + { + "epoch": 0.0205213397503237, + "grad_norm": 0.006109077949076891, + "learning_rate": 0.2999187861912387, + "loss": 0.1224, + "num_input_tokens_seen": 546464, + "step": 420 + }, + { + "epoch": 0.02076564141401803, + "grad_norm": 0.015186105854809284, + "learning_rate": 0.2999168365294737, + "loss": 0.1575, + "num_input_tokens_seen": 552928, + "step": 425 + }, + { + "epoch": 0.02100994307771236, + "grad_norm": 0.02595384046435356, + "learning_rate": 0.29991486374864856, + "loss": 0.1603, + "num_input_tokens_seen": 559168, + "step": 430 + }, + { + "epoch": 0.021254244741406688, + "grad_norm": 0.01085552666336298, + "learning_rate": 0.29991286784906745, + "loss": 0.1627, + "num_input_tokens_seen": 565312, + "step": 435 + }, + { + "epoch": 0.02149854640510102, + "grad_norm": 0.007158820983022451, + "learning_rate": 0.2999108488310382, + "loss": 0.1524, + "num_input_tokens_seen": 571488, + "step": 440 + }, + { + "epoch": 0.02174284806879535, + "grad_norm": 0.011095430701971054, + "learning_rate": 0.29990880669487213, + "loss": 0.1624, + "num_input_tokens_seen": 578016, + "step": 445 + }, + { + "epoch": 0.021987149732489678, + "grad_norm": 0.012175707146525383, + "learning_rate": 0.29990674144088425, + "loss": 0.1407, + "num_input_tokens_seen": 584480, + "step": 450 + }, + { + "epoch": 0.02223145139618401, + "grad_norm": 0.016517313197255135, + "learning_rate": 0.299904653069393, + "loss": 0.157, + "num_input_tokens_seen": 591648, + "step": 455 + }, + { + "epoch": 0.02247575305987834, + "grad_norm": 0.021303463727235794, + "learning_rate": 0.29990254158072044, + "loss": 0.1711, + "num_input_tokens_seen": 597760, + "step": 460 + }, + { + "epoch": 0.022720054723572667, + "grad_norm": 0.011815404519438744, + "learning_rate": 0.2999004069751921, + "loss": 0.168, + "num_input_tokens_seen": 604320, + "step": 465 + }, + { + "epoch": 0.022964356387266998, + "grad_norm": 0.021556803956627846, + "learning_rate": 0.2998982492531373, + "loss": 0.1392, + "num_input_tokens_seen": 610592, + "step": 470 + }, + { + "epoch": 0.023208658050961326, + "grad_norm": 0.007045431528240442, + "learning_rate": 0.2998960684148887, + "loss": 0.1494, + "num_input_tokens_seen": 617120, + "step": 475 + }, + { + "epoch": 0.023452959714655656, + "grad_norm": 0.007029450964182615, + "learning_rate": 0.29989386446078264, + "loss": 0.1698, + "num_input_tokens_seen": 623296, + "step": 480 + }, + { + "epoch": 0.023697261378349987, + "grad_norm": 0.0161809753626585, + "learning_rate": 0.299891637391159, + "loss": 0.1465, + "num_input_tokens_seen": 629984, + "step": 485 + }, + { + "epoch": 0.023941563042044315, + "grad_norm": 0.017255816608667374, + "learning_rate": 0.2998893872063612, + "loss": 0.1417, + "num_input_tokens_seen": 636800, + "step": 490 + }, + { + "epoch": 0.024185864705738646, + "grad_norm": 0.007997923530638218, + "learning_rate": 0.2998871139067363, + "loss": 0.1558, + "num_input_tokens_seen": 643456, + "step": 495 + }, + { + "epoch": 0.024430166369432977, + "grad_norm": 0.016810359433293343, + "learning_rate": 0.2998848174926348, + "loss": 0.1552, + "num_input_tokens_seen": 650720, + "step": 500 + }, + { + "epoch": 0.024674468033127304, + "grad_norm": 0.013938031159341335, + "learning_rate": 0.2998824979644109, + "loss": 0.1611, + "num_input_tokens_seen": 657440, + "step": 505 + }, + { + "epoch": 0.024918769696821635, + "grad_norm": 0.006744832266122103, + "learning_rate": 0.29988015532242224, + "loss": 0.1492, + "num_input_tokens_seen": 663680, + "step": 510 + }, + { + "epoch": 0.025163071360515966, + "grad_norm": 0.02286100760102272, + "learning_rate": 0.29987778956703015, + "loss": 0.1589, + "num_input_tokens_seen": 670784, + "step": 515 + }, + { + "epoch": 0.025407373024210294, + "grad_norm": 0.009795085527002811, + "learning_rate": 0.2998754006985994, + "loss": 0.1519, + "num_input_tokens_seen": 677760, + "step": 520 + }, + { + "epoch": 0.025651674687904625, + "grad_norm": 0.009243443608283997, + "learning_rate": 0.29987298871749846, + "loss": 0.152, + "num_input_tokens_seen": 684192, + "step": 525 + }, + { + "epoch": 0.025895976351598956, + "grad_norm": 0.005852101370692253, + "learning_rate": 0.2998705536240992, + "loss": 0.156, + "num_input_tokens_seen": 690432, + "step": 530 + }, + { + "epoch": 0.026140278015293283, + "grad_norm": 0.009896044619381428, + "learning_rate": 0.2998680954187772, + "loss": 0.1709, + "num_input_tokens_seen": 696800, + "step": 535 + }, + { + "epoch": 0.026384579678987614, + "grad_norm": 0.008009972050786018, + "learning_rate": 0.2998656141019115, + "loss": 0.1551, + "num_input_tokens_seen": 703392, + "step": 540 + }, + { + "epoch": 0.026628881342681942, + "grad_norm": 0.005669897887855768, + "learning_rate": 0.2998631096738848, + "loss": 0.1486, + "num_input_tokens_seen": 709728, + "step": 545 + }, + { + "epoch": 0.026873183006376273, + "grad_norm": 0.010589969344437122, + "learning_rate": 0.29986058213508326, + "loss": 0.1593, + "num_input_tokens_seen": 715904, + "step": 550 + }, + { + "epoch": 0.027117484670070604, + "grad_norm": 0.013816055841743946, + "learning_rate": 0.29985803148589674, + "loss": 0.1691, + "num_input_tokens_seen": 722304, + "step": 555 + }, + { + "epoch": 0.02736178633376493, + "grad_norm": 0.006446806248277426, + "learning_rate": 0.2998554577267185, + "loss": 0.1656, + "num_input_tokens_seen": 728320, + "step": 560 + }, + { + "epoch": 0.027606087997459262, + "grad_norm": 0.010421511717140675, + "learning_rate": 0.2998528608579455, + "loss": 0.1739, + "num_input_tokens_seen": 734496, + "step": 565 + }, + { + "epoch": 0.027850389661153593, + "grad_norm": 0.015989653766155243, + "learning_rate": 0.2998502408799781, + "loss": 0.1485, + "num_input_tokens_seen": 741248, + "step": 570 + }, + { + "epoch": 0.02809469132484792, + "grad_norm": 0.010897301137447357, + "learning_rate": 0.2998475977932205, + "loss": 0.1775, + "num_input_tokens_seen": 747712, + "step": 575 + }, + { + "epoch": 0.028338992988542252, + "grad_norm": 0.00813302118331194, + "learning_rate": 0.29984493159808023, + "loss": 0.1391, + "num_input_tokens_seen": 755040, + "step": 580 + }, + { + "epoch": 0.028583294652236583, + "grad_norm": 0.022107519209384918, + "learning_rate": 0.29984224229496836, + "loss": 0.1553, + "num_input_tokens_seen": 761536, + "step": 585 + }, + { + "epoch": 0.02882759631593091, + "grad_norm": 0.01644737832248211, + "learning_rate": 0.2998395298842998, + "loss": 0.1442, + "num_input_tokens_seen": 768000, + "step": 590 + }, + { + "epoch": 0.02907189797962524, + "grad_norm": 0.005235717631876469, + "learning_rate": 0.29983679436649263, + "loss": 0.1452, + "num_input_tokens_seen": 774112, + "step": 595 + }, + { + "epoch": 0.029316199643319572, + "grad_norm": 0.005933593492954969, + "learning_rate": 0.2998340357419689, + "loss": 0.1369, + "num_input_tokens_seen": 780768, + "step": 600 + }, + { + "epoch": 0.029316199643319572, + "eval_loss": 0.17752529680728912, + "eval_runtime": 402.1706, + "eval_samples_per_second": 90.472, + "eval_steps_per_second": 22.62, + "num_input_tokens_seen": 780768, + "step": 600 + }, + { + "epoch": 0.0295605013070139, + "grad_norm": 0.008518454618752003, + "learning_rate": 0.29983125401115385, + "loss": 0.168, + "num_input_tokens_seen": 787104, + "step": 605 + }, + { + "epoch": 0.02980480297070823, + "grad_norm": 0.010880297049880028, + "learning_rate": 0.29982844917447654, + "loss": 0.1616, + "num_input_tokens_seen": 793856, + "step": 610 + }, + { + "epoch": 0.030049104634402562, + "grad_norm": 0.012697809375822544, + "learning_rate": 0.2998256212323695, + "loss": 0.1663, + "num_input_tokens_seen": 800448, + "step": 615 + }, + { + "epoch": 0.03029340629809689, + "grad_norm": 0.007531088311225176, + "learning_rate": 0.29982277018526887, + "loss": 0.1372, + "num_input_tokens_seen": 807232, + "step": 620 + }, + { + "epoch": 0.03053770796179122, + "grad_norm": 0.00527376402169466, + "learning_rate": 0.2998198960336143, + "loss": 0.1585, + "num_input_tokens_seen": 813600, + "step": 625 + }, + { + "epoch": 0.030782009625485548, + "grad_norm": 0.004967930726706982, + "learning_rate": 0.299816998777849, + "loss": 0.1336, + "num_input_tokens_seen": 820000, + "step": 630 + }, + { + "epoch": 0.03102631128917988, + "grad_norm": 0.006969386711716652, + "learning_rate": 0.2998140784184197, + "loss": 0.1406, + "num_input_tokens_seen": 826560, + "step": 635 + }, + { + "epoch": 0.031270612952874206, + "grad_norm": 0.012204078957438469, + "learning_rate": 0.2998111349557769, + "loss": 0.1418, + "num_input_tokens_seen": 832960, + "step": 640 + }, + { + "epoch": 0.03151491461656854, + "grad_norm": 0.011341945268213749, + "learning_rate": 0.29980816839037444, + "loss": 0.1447, + "num_input_tokens_seen": 839680, + "step": 645 + }, + { + "epoch": 0.03175921628026287, + "grad_norm": 0.006339394953101873, + "learning_rate": 0.2998051787226698, + "loss": 0.1412, + "num_input_tokens_seen": 846592, + "step": 650 + }, + { + "epoch": 0.0320035179439572, + "grad_norm": 0.009623605757951736, + "learning_rate": 0.29980216595312403, + "loss": 0.139, + "num_input_tokens_seen": 853088, + "step": 655 + }, + { + "epoch": 0.03224781960765153, + "grad_norm": 0.011906609870493412, + "learning_rate": 0.29979913008220177, + "loss": 0.1546, + "num_input_tokens_seen": 859424, + "step": 660 + }, + { + "epoch": 0.03249212127134586, + "grad_norm": 0.014531293883919716, + "learning_rate": 0.2997960711103711, + "loss": 0.1418, + "num_input_tokens_seen": 866080, + "step": 665 + }, + { + "epoch": 0.032736422935040185, + "grad_norm": 0.011538858525454998, + "learning_rate": 0.29979298903810386, + "loss": 0.1175, + "num_input_tokens_seen": 872608, + "step": 670 + }, + { + "epoch": 0.032980724598734516, + "grad_norm": 0.007460369728505611, + "learning_rate": 0.29978988386587524, + "loss": 0.1655, + "num_input_tokens_seen": 879008, + "step": 675 + }, + { + "epoch": 0.03322502626242885, + "grad_norm": 0.0071784802712500095, + "learning_rate": 0.2997867555941642, + "loss": 0.1516, + "num_input_tokens_seen": 885408, + "step": 680 + }, + { + "epoch": 0.03346932792612318, + "grad_norm": 0.006322260946035385, + "learning_rate": 0.299783604223453, + "loss": 0.1506, + "num_input_tokens_seen": 891552, + "step": 685 + }, + { + "epoch": 0.03371362958981751, + "grad_norm": 0.012253022752702236, + "learning_rate": 0.29978042975422786, + "loss": 0.1426, + "num_input_tokens_seen": 897984, + "step": 690 + }, + { + "epoch": 0.03395793125351183, + "grad_norm": 0.013147941790521145, + "learning_rate": 0.29977723218697816, + "loss": 0.1595, + "num_input_tokens_seen": 904480, + "step": 695 + }, + { + "epoch": 0.034202232917206164, + "grad_norm": 0.00795318465679884, + "learning_rate": 0.299774011522197, + "loss": 0.1384, + "num_input_tokens_seen": 910880, + "step": 700 + }, + { + "epoch": 0.034446534580900495, + "grad_norm": 0.015226255171000957, + "learning_rate": 0.29977076776038114, + "loss": 0.1657, + "num_input_tokens_seen": 917216, + "step": 705 + }, + { + "epoch": 0.034690836244594826, + "grad_norm": 0.012421595863997936, + "learning_rate": 0.2997675009020307, + "loss": 0.1444, + "num_input_tokens_seen": 923232, + "step": 710 + }, + { + "epoch": 0.03493513790828916, + "grad_norm": 0.00955990795046091, + "learning_rate": 0.2997642109476496, + "loss": 0.1635, + "num_input_tokens_seen": 929824, + "step": 715 + }, + { + "epoch": 0.03517943957198349, + "grad_norm": 0.006837652996182442, + "learning_rate": 0.299760897897745, + "loss": 0.1574, + "num_input_tokens_seen": 936224, + "step": 720 + }, + { + "epoch": 0.03542374123567781, + "grad_norm": 0.005699640605598688, + "learning_rate": 0.29975756175282803, + "loss": 0.1421, + "num_input_tokens_seen": 942432, + "step": 725 + }, + { + "epoch": 0.03566804289937214, + "grad_norm": 0.006488234270364046, + "learning_rate": 0.29975420251341306, + "loss": 0.1544, + "num_input_tokens_seen": 948672, + "step": 730 + }, + { + "epoch": 0.035912344563066474, + "grad_norm": 0.00888883788138628, + "learning_rate": 0.29975082018001814, + "loss": 0.1687, + "num_input_tokens_seen": 954880, + "step": 735 + }, + { + "epoch": 0.036156646226760805, + "grad_norm": 0.015939664095640182, + "learning_rate": 0.2997474147531648, + "loss": 0.1702, + "num_input_tokens_seen": 961120, + "step": 740 + }, + { + "epoch": 0.036400947890455136, + "grad_norm": 0.007008249871432781, + "learning_rate": 0.29974398623337833, + "loss": 0.1599, + "num_input_tokens_seen": 967616, + "step": 745 + }, + { + "epoch": 0.03664524955414947, + "grad_norm": 0.007346251513808966, + "learning_rate": 0.2997405346211873, + "loss": 0.1567, + "num_input_tokens_seen": 974272, + "step": 750 + }, + { + "epoch": 0.03688955121784379, + "grad_norm": 0.007338172756135464, + "learning_rate": 0.2997370599171241, + "loss": 0.1468, + "num_input_tokens_seen": 980608, + "step": 755 + }, + { + "epoch": 0.03713385288153812, + "grad_norm": 0.00700546707957983, + "learning_rate": 0.2997335621217246, + "loss": 0.141, + "num_input_tokens_seen": 987168, + "step": 760 + }, + { + "epoch": 0.03737815454523245, + "grad_norm": 0.016759589314460754, + "learning_rate": 0.29973004123552816, + "loss": 0.1408, + "num_input_tokens_seen": 993632, + "step": 765 + }, + { + "epoch": 0.037622456208926784, + "grad_norm": 0.006739115342497826, + "learning_rate": 0.2997264972590777, + "loss": 0.1594, + "num_input_tokens_seen": 999968, + "step": 770 + }, + { + "epoch": 0.037866757872621115, + "grad_norm": 0.006750429980456829, + "learning_rate": 0.29972293019291973, + "loss": 0.1464, + "num_input_tokens_seen": 1006688, + "step": 775 + }, + { + "epoch": 0.03811105953631544, + "grad_norm": 0.028798270970582962, + "learning_rate": 0.2997193400376045, + "loss": 0.1663, + "num_input_tokens_seen": 1012896, + "step": 780 + }, + { + "epoch": 0.03835536120000977, + "grad_norm": 0.01769954338669777, + "learning_rate": 0.2997157267936854, + "loss": 0.1677, + "num_input_tokens_seen": 1019136, + "step": 785 + }, + { + "epoch": 0.0385996628637041, + "grad_norm": 0.0055108931846916676, + "learning_rate": 0.2997120904617199, + "loss": 0.1456, + "num_input_tokens_seen": 1025792, + "step": 790 + }, + { + "epoch": 0.03884396452739843, + "grad_norm": 0.00925780925899744, + "learning_rate": 0.29970843104226863, + "loss": 0.1609, + "num_input_tokens_seen": 1032032, + "step": 795 + }, + { + "epoch": 0.03908826619109276, + "grad_norm": 0.009362381882965565, + "learning_rate": 0.2997047485358959, + "loss": 0.1529, + "num_input_tokens_seen": 1038304, + "step": 800 + }, + { + "epoch": 0.03908826619109276, + "eval_loss": 0.14968900382518768, + "eval_runtime": 401.9259, + "eval_samples_per_second": 90.527, + "eval_steps_per_second": 22.634, + "num_input_tokens_seen": 1038304, + "step": 800 + }, + { + "epoch": 0.039332567854787094, + "grad_norm": 0.008691994473338127, + "learning_rate": 0.2997010429431697, + "loss": 0.1707, + "num_input_tokens_seen": 1044384, + "step": 805 + }, + { + "epoch": 0.03957686951848142, + "grad_norm": 0.009997324086725712, + "learning_rate": 0.29969731426466134, + "loss": 0.1701, + "num_input_tokens_seen": 1051040, + "step": 810 + }, + { + "epoch": 0.03982117118217575, + "grad_norm": 0.005586818791925907, + "learning_rate": 0.299693562500946, + "loss": 0.1391, + "num_input_tokens_seen": 1057664, + "step": 815 + }, + { + "epoch": 0.04006547284587008, + "grad_norm": 0.008421546779572964, + "learning_rate": 0.29968978765260207, + "loss": 0.1455, + "num_input_tokens_seen": 1063968, + "step": 820 + }, + { + "epoch": 0.04030977450956441, + "grad_norm": 0.006749195046722889, + "learning_rate": 0.2996859897202118, + "loss": 0.1613, + "num_input_tokens_seen": 1070624, + "step": 825 + }, + { + "epoch": 0.04055407617325874, + "grad_norm": 0.011877949349582195, + "learning_rate": 0.2996821687043609, + "loss": 0.1442, + "num_input_tokens_seen": 1077376, + "step": 830 + }, + { + "epoch": 0.04079837783695307, + "grad_norm": 0.01245996356010437, + "learning_rate": 0.2996783246056384, + "loss": 0.1478, + "num_input_tokens_seen": 1083840, + "step": 835 + }, + { + "epoch": 0.0410426795006474, + "grad_norm": 0.009998328052461147, + "learning_rate": 0.29967445742463744, + "loss": 0.1647, + "num_input_tokens_seen": 1090272, + "step": 840 + }, + { + "epoch": 0.04128698116434173, + "grad_norm": 0.005372264422476292, + "learning_rate": 0.29967056716195417, + "loss": 0.1312, + "num_input_tokens_seen": 1096992, + "step": 845 + }, + { + "epoch": 0.04153128282803606, + "grad_norm": 0.016267532482743263, + "learning_rate": 0.2996666538181885, + "loss": 0.1957, + "num_input_tokens_seen": 1103392, + "step": 850 + }, + { + "epoch": 0.04177558449173039, + "grad_norm": 0.011358632706105709, + "learning_rate": 0.29966271739394407, + "loss": 0.1851, + "num_input_tokens_seen": 1109664, + "step": 855 + }, + { + "epoch": 0.04201988615542472, + "grad_norm": 0.00457138242200017, + "learning_rate": 0.29965875788982776, + "loss": 0.1513, + "num_input_tokens_seen": 1116096, + "step": 860 + }, + { + "epoch": 0.042264187819119045, + "grad_norm": 0.008696872740983963, + "learning_rate": 0.2996547753064503, + "loss": 0.1472, + "num_input_tokens_seen": 1122048, + "step": 865 + }, + { + "epoch": 0.042508489482813376, + "grad_norm": 0.008294257335364819, + "learning_rate": 0.29965076964442583, + "loss": 0.1435, + "num_input_tokens_seen": 1128512, + "step": 870 + }, + { + "epoch": 0.04275279114650771, + "grad_norm": 0.008795199915766716, + "learning_rate": 0.299646740904372, + "loss": 0.1448, + "num_input_tokens_seen": 1134912, + "step": 875 + }, + { + "epoch": 0.04299709281020204, + "grad_norm": 0.01687173917889595, + "learning_rate": 0.29964268908691016, + "loss": 0.1375, + "num_input_tokens_seen": 1141504, + "step": 880 + }, + { + "epoch": 0.04324139447389637, + "grad_norm": 0.01937958225607872, + "learning_rate": 0.29963861419266513, + "loss": 0.1634, + "num_input_tokens_seen": 1147872, + "step": 885 + }, + { + "epoch": 0.0434856961375907, + "grad_norm": 0.008877787739038467, + "learning_rate": 0.29963451622226533, + "loss": 0.1368, + "num_input_tokens_seen": 1154848, + "step": 890 + }, + { + "epoch": 0.043729997801285024, + "grad_norm": 0.009217260405421257, + "learning_rate": 0.29963039517634277, + "loss": 0.1281, + "num_input_tokens_seen": 1161248, + "step": 895 + }, + { + "epoch": 0.043974299464979355, + "grad_norm": 0.014786854386329651, + "learning_rate": 0.2996262510555328, + "loss": 0.1509, + "num_input_tokens_seen": 1168032, + "step": 900 + }, + { + "epoch": 0.044218601128673686, + "grad_norm": 0.010081278160214424, + "learning_rate": 0.2996220838604746, + "loss": 0.1489, + "num_input_tokens_seen": 1174304, + "step": 905 + }, + { + "epoch": 0.04446290279236802, + "grad_norm": 0.02486228384077549, + "learning_rate": 0.29961789359181085, + "loss": 0.1634, + "num_input_tokens_seen": 1180640, + "step": 910 + }, + { + "epoch": 0.04470720445606235, + "grad_norm": 0.007458200212568045, + "learning_rate": 0.29961368025018764, + "loss": 0.1667, + "num_input_tokens_seen": 1187104, + "step": 915 + }, + { + "epoch": 0.04495150611975668, + "grad_norm": 0.023550761863589287, + "learning_rate": 0.2996094438362548, + "loss": 0.1756, + "num_input_tokens_seen": 1193600, + "step": 920 + }, + { + "epoch": 0.045195807783451, + "grad_norm": 0.006629835814237595, + "learning_rate": 0.2996051843506657, + "loss": 0.1406, + "num_input_tokens_seen": 1200384, + "step": 925 + }, + { + "epoch": 0.045440109447145334, + "grad_norm": 0.01368903648108244, + "learning_rate": 0.299600901794077, + "loss": 0.1617, + "num_input_tokens_seen": 1206528, + "step": 930 + }, + { + "epoch": 0.045684411110839665, + "grad_norm": 0.012660546228289604, + "learning_rate": 0.29959659616714923, + "loss": 0.1374, + "num_input_tokens_seen": 1212800, + "step": 935 + }, + { + "epoch": 0.045928712774533996, + "grad_norm": 0.012615502811968327, + "learning_rate": 0.2995922674705464, + "loss": 0.1461, + "num_input_tokens_seen": 1218976, + "step": 940 + }, + { + "epoch": 0.04617301443822833, + "grad_norm": 0.02121579460799694, + "learning_rate": 0.2995879157049361, + "loss": 0.1605, + "num_input_tokens_seen": 1225376, + "step": 945 + }, + { + "epoch": 0.04641731610192265, + "grad_norm": 0.008519984781742096, + "learning_rate": 0.2995835408709893, + "loss": 0.1461, + "num_input_tokens_seen": 1231360, + "step": 950 + }, + { + "epoch": 0.04666161776561698, + "grad_norm": 0.01731475256383419, + "learning_rate": 0.29957914296938076, + "loss": 0.1632, + "num_input_tokens_seen": 1237856, + "step": 955 + }, + { + "epoch": 0.04690591942931131, + "grad_norm": 0.0149699617177248, + "learning_rate": 0.2995747220007886, + "loss": 0.162, + "num_input_tokens_seen": 1244384, + "step": 960 + }, + { + "epoch": 0.047150221093005644, + "grad_norm": 0.012257126159965992, + "learning_rate": 0.2995702779658947, + "loss": 0.1451, + "num_input_tokens_seen": 1250816, + "step": 965 + }, + { + "epoch": 0.047394522756699975, + "grad_norm": 0.008137860335409641, + "learning_rate": 0.29956581086538425, + "loss": 0.1287, + "num_input_tokens_seen": 1257504, + "step": 970 + }, + { + "epoch": 0.047638824420394306, + "grad_norm": 0.015022856183350086, + "learning_rate": 0.2995613206999462, + "loss": 0.1354, + "num_input_tokens_seen": 1264160, + "step": 975 + }, + { + "epoch": 0.04788312608408863, + "grad_norm": 0.019941039383411407, + "learning_rate": 0.29955680747027297, + "loss": 0.1461, + "num_input_tokens_seen": 1271008, + "step": 980 + }, + { + "epoch": 0.04812742774778296, + "grad_norm": 0.016780568286776543, + "learning_rate": 0.2995522711770607, + "loss": 0.1695, + "num_input_tokens_seen": 1277248, + "step": 985 + }, + { + "epoch": 0.04837172941147729, + "grad_norm": 0.008415346965193748, + "learning_rate": 0.2995477118210087, + "loss": 0.1549, + "num_input_tokens_seen": 1283712, + "step": 990 + }, + { + "epoch": 0.04861603107517162, + "grad_norm": 0.013504128903150558, + "learning_rate": 0.29954312940282024, + "loss": 0.1601, + "num_input_tokens_seen": 1289632, + "step": 995 + }, + { + "epoch": 0.048860332738865954, + "grad_norm": 0.012223360128700733, + "learning_rate": 0.29953852392320196, + "loss": 0.1607, + "num_input_tokens_seen": 1296288, + "step": 1000 + }, + { + "epoch": 0.048860332738865954, + "eval_loss": 0.1531103551387787, + "eval_runtime": 402.1334, + "eval_samples_per_second": 90.48, + "eval_steps_per_second": 22.622, + "num_input_tokens_seen": 1296288, + "step": 1000 + }, + { + "epoch": 0.049104634402560285, + "grad_norm": 0.00928540714085102, + "learning_rate": 0.2995338953828641, + "loss": 0.1363, + "num_input_tokens_seen": 1302752, + "step": 1005 + }, + { + "epoch": 0.04934893606625461, + "grad_norm": 0.011860625818371773, + "learning_rate": 0.2995292437825204, + "loss": 0.1396, + "num_input_tokens_seen": 1309152, + "step": 1010 + }, + { + "epoch": 0.04959323772994894, + "grad_norm": 0.006690794602036476, + "learning_rate": 0.29952456912288816, + "loss": 0.145, + "num_input_tokens_seen": 1315584, + "step": 1015 + }, + { + "epoch": 0.04983753939364327, + "grad_norm": 0.01082373782992363, + "learning_rate": 0.2995198714046884, + "loss": 0.1439, + "num_input_tokens_seen": 1322368, + "step": 1020 + }, + { + "epoch": 0.0500818410573376, + "grad_norm": 0.010472146794199944, + "learning_rate": 0.2995151506286454, + "loss": 0.1334, + "num_input_tokens_seen": 1329152, + "step": 1025 + }, + { + "epoch": 0.05032614272103193, + "grad_norm": 0.015635352581739426, + "learning_rate": 0.2995104067954873, + "loss": 0.1779, + "num_input_tokens_seen": 1335712, + "step": 1030 + }, + { + "epoch": 0.05057044438472626, + "grad_norm": 0.011368009261786938, + "learning_rate": 0.2995056399059456, + "loss": 0.1671, + "num_input_tokens_seen": 1341632, + "step": 1035 + }, + { + "epoch": 0.05081474604842059, + "grad_norm": 0.008341777138411999, + "learning_rate": 0.2995008499607554, + "loss": 0.1645, + "num_input_tokens_seen": 1348096, + "step": 1040 + }, + { + "epoch": 0.05105904771211492, + "grad_norm": 0.007754199206829071, + "learning_rate": 0.2994960369606554, + "loss": 0.1583, + "num_input_tokens_seen": 1354432, + "step": 1045 + }, + { + "epoch": 0.05130334937580925, + "grad_norm": 0.009154225699603558, + "learning_rate": 0.2994912009063878, + "loss": 0.1614, + "num_input_tokens_seen": 1360864, + "step": 1050 + }, + { + "epoch": 0.05154765103950358, + "grad_norm": 0.013811626471579075, + "learning_rate": 0.29948634179869843, + "loss": 0.1663, + "num_input_tokens_seen": 1366848, + "step": 1055 + }, + { + "epoch": 0.05179195270319791, + "grad_norm": 0.007183332461863756, + "learning_rate": 0.29948145963833656, + "loss": 0.1597, + "num_input_tokens_seen": 1372960, + "step": 1060 + }, + { + "epoch": 0.052036254366892236, + "grad_norm": 0.008194766007363796, + "learning_rate": 0.29947655442605514, + "loss": 0.1695, + "num_input_tokens_seen": 1380032, + "step": 1065 + }, + { + "epoch": 0.05228055603058657, + "grad_norm": 0.0037348486948758364, + "learning_rate": 0.2994716261626106, + "loss": 0.1381, + "num_input_tokens_seen": 1386464, + "step": 1070 + }, + { + "epoch": 0.0525248576942809, + "grad_norm": 0.005621443968266249, + "learning_rate": 0.2994666748487629, + "loss": 0.1601, + "num_input_tokens_seen": 1392992, + "step": 1075 + }, + { + "epoch": 0.05276915935797523, + "grad_norm": 0.005465729162096977, + "learning_rate": 0.2994617004852756, + "loss": 0.148, + "num_input_tokens_seen": 1399360, + "step": 1080 + }, + { + "epoch": 0.05301346102166956, + "grad_norm": 0.007075119763612747, + "learning_rate": 0.2994567030729159, + "loss": 0.1554, + "num_input_tokens_seen": 1405760, + "step": 1085 + }, + { + "epoch": 0.053257762685363884, + "grad_norm": 0.015542961657047272, + "learning_rate": 0.29945168261245436, + "loss": 0.158, + "num_input_tokens_seen": 1412000, + "step": 1090 + }, + { + "epoch": 0.053502064349058215, + "grad_norm": 0.011317209340631962, + "learning_rate": 0.29944663910466524, + "loss": 0.1485, + "num_input_tokens_seen": 1418240, + "step": 1095 + }, + { + "epoch": 0.053746366012752546, + "grad_norm": 0.00800265558063984, + "learning_rate": 0.2994415725503263, + "loss": 0.1528, + "num_input_tokens_seen": 1424576, + "step": 1100 + }, + { + "epoch": 0.05399066767644688, + "grad_norm": 0.013979986310005188, + "learning_rate": 0.29943648295021885, + "loss": 0.1616, + "num_input_tokens_seen": 1430720, + "step": 1105 + }, + { + "epoch": 0.05423496934014121, + "grad_norm": 0.0063639115542173386, + "learning_rate": 0.2994313703051278, + "loss": 0.1356, + "num_input_tokens_seen": 1437280, + "step": 1110 + }, + { + "epoch": 0.05447927100383554, + "grad_norm": 0.009362876415252686, + "learning_rate": 0.29942623461584156, + "loss": 0.1446, + "num_input_tokens_seen": 1443872, + "step": 1115 + }, + { + "epoch": 0.05472357266752986, + "grad_norm": 0.005186263006180525, + "learning_rate": 0.29942107588315214, + "loss": 0.1348, + "num_input_tokens_seen": 1450304, + "step": 1120 + }, + { + "epoch": 0.054967874331224194, + "grad_norm": 0.01479154173284769, + "learning_rate": 0.29941589410785513, + "loss": 0.1592, + "num_input_tokens_seen": 1456288, + "step": 1125 + }, + { + "epoch": 0.055212175994918525, + "grad_norm": 0.00915540661662817, + "learning_rate": 0.29941068929074954, + "loss": 0.1536, + "num_input_tokens_seen": 1462944, + "step": 1130 + }, + { + "epoch": 0.055456477658612856, + "grad_norm": 0.007621387951076031, + "learning_rate": 0.2994054614326381, + "loss": 0.1433, + "num_input_tokens_seen": 1469472, + "step": 1135 + }, + { + "epoch": 0.05570077932230719, + "grad_norm": 0.005959915928542614, + "learning_rate": 0.29940021053432686, + "loss": 0.1486, + "num_input_tokens_seen": 1475648, + "step": 1140 + }, + { + "epoch": 0.05594508098600152, + "grad_norm": 0.007811995688825846, + "learning_rate": 0.29939493659662575, + "loss": 0.1531, + "num_input_tokens_seen": 1482432, + "step": 1145 + }, + { + "epoch": 0.05618938264969584, + "grad_norm": 0.006437181029468775, + "learning_rate": 0.299389639620348, + "loss": 0.1493, + "num_input_tokens_seen": 1489152, + "step": 1150 + }, + { + "epoch": 0.05643368431339017, + "grad_norm": 0.009278996847569942, + "learning_rate": 0.29938431960631046, + "loss": 0.1341, + "num_input_tokens_seen": 1495232, + "step": 1155 + }, + { + "epoch": 0.056677985977084504, + "grad_norm": 0.013540171086788177, + "learning_rate": 0.2993789765553335, + "loss": 0.1622, + "num_input_tokens_seen": 1502176, + "step": 1160 + }, + { + "epoch": 0.056922287640778835, + "grad_norm": 0.005356914363801479, + "learning_rate": 0.2993736104682412, + "loss": 0.1357, + "num_input_tokens_seen": 1509408, + "step": 1165 + }, + { + "epoch": 0.057166589304473166, + "grad_norm": 0.013111709617078304, + "learning_rate": 0.299368221345861, + "loss": 0.1757, + "num_input_tokens_seen": 1515808, + "step": 1170 + }, + { + "epoch": 0.05741089096816749, + "grad_norm": 0.016909131780266762, + "learning_rate": 0.29936280918902397, + "loss": 0.1799, + "num_input_tokens_seen": 1521856, + "step": 1175 + }, + { + "epoch": 0.05765519263186182, + "grad_norm": 0.006668919697403908, + "learning_rate": 0.2993573739985648, + "loss": 0.1555, + "num_input_tokens_seen": 1528000, + "step": 1180 + }, + { + "epoch": 0.05789949429555615, + "grad_norm": 0.007245638407766819, + "learning_rate": 0.2993519157753216, + "loss": 0.1377, + "num_input_tokens_seen": 1534816, + "step": 1185 + }, + { + "epoch": 0.05814379595925048, + "grad_norm": 0.009512417949736118, + "learning_rate": 0.2993464345201361, + "loss": 0.1517, + "num_input_tokens_seen": 1541568, + "step": 1190 + }, + { + "epoch": 0.058388097622944814, + "grad_norm": 0.009734654799103737, + "learning_rate": 0.2993409302338536, + "loss": 0.1657, + "num_input_tokens_seen": 1547680, + "step": 1195 + }, + { + "epoch": 0.058632399286639145, + "grad_norm": 0.007397621404379606, + "learning_rate": 0.2993354029173229, + "loss": 0.1522, + "num_input_tokens_seen": 1554400, + "step": 1200 + }, + { + "epoch": 0.058632399286639145, + "eval_loss": 0.17315985262393951, + "eval_runtime": 402.4413, + "eval_samples_per_second": 90.411, + "eval_steps_per_second": 22.605, + "num_input_tokens_seen": 1554400, + "step": 1200 + }, + { + "epoch": 0.05887670095033347, + "grad_norm": 0.011762763373553753, + "learning_rate": 0.2993298525713965, + "loss": 0.1739, + "num_input_tokens_seen": 1560960, + "step": 1205 + }, + { + "epoch": 0.0591210026140278, + "grad_norm": 0.004923292435705662, + "learning_rate": 0.29932427919693017, + "loss": 0.1607, + "num_input_tokens_seen": 1567648, + "step": 1210 + }, + { + "epoch": 0.05936530427772213, + "grad_norm": 0.0059710219502449036, + "learning_rate": 0.2993186827947834, + "loss": 0.1601, + "num_input_tokens_seen": 1573728, + "step": 1215 + }, + { + "epoch": 0.05960960594141646, + "grad_norm": 0.005825436674058437, + "learning_rate": 0.2993130633658194, + "loss": 0.1509, + "num_input_tokens_seen": 1580416, + "step": 1220 + }, + { + "epoch": 0.05985390760511079, + "grad_norm": 0.007773267105221748, + "learning_rate": 0.29930742091090456, + "loss": 0.144, + "num_input_tokens_seen": 1586816, + "step": 1225 + }, + { + "epoch": 0.060098209268805124, + "grad_norm": 0.009800752624869347, + "learning_rate": 0.29930175543090914, + "loss": 0.1401, + "num_input_tokens_seen": 1593344, + "step": 1230 + }, + { + "epoch": 0.06034251093249945, + "grad_norm": 0.007993437349796295, + "learning_rate": 0.2992960669267068, + "loss": 0.1307, + "num_input_tokens_seen": 1599648, + "step": 1235 + }, + { + "epoch": 0.06058681259619378, + "grad_norm": 0.00866819079965353, + "learning_rate": 0.29929035539917476, + "loss": 0.1631, + "num_input_tokens_seen": 1606144, + "step": 1240 + }, + { + "epoch": 0.06083111425988811, + "grad_norm": 0.009353644214570522, + "learning_rate": 0.2992846208491938, + "loss": 0.1403, + "num_input_tokens_seen": 1612640, + "step": 1245 + }, + { + "epoch": 0.06107541592358244, + "grad_norm": 0.01155051402747631, + "learning_rate": 0.2992788632776483, + "loss": 0.1405, + "num_input_tokens_seen": 1619040, + "step": 1250 + }, + { + "epoch": 0.06131971758727677, + "grad_norm": 0.008947743102908134, + "learning_rate": 0.29927308268542613, + "loss": 0.1491, + "num_input_tokens_seen": 1625184, + "step": 1255 + }, + { + "epoch": 0.061564019250971096, + "grad_norm": 0.009883292019367218, + "learning_rate": 0.2992672790734187, + "loss": 0.1529, + "num_input_tokens_seen": 1631520, + "step": 1260 + }, + { + "epoch": 0.06180832091466543, + "grad_norm": 0.004838760010898113, + "learning_rate": 0.299261452442521, + "loss": 0.1489, + "num_input_tokens_seen": 1638080, + "step": 1265 + }, + { + "epoch": 0.06205262257835976, + "grad_norm": 0.006850129459053278, + "learning_rate": 0.29925560279363167, + "loss": 0.1504, + "num_input_tokens_seen": 1644384, + "step": 1270 + }, + { + "epoch": 0.06229692424205409, + "grad_norm": 0.0063577136024832726, + "learning_rate": 0.29924973012765266, + "loss": 0.149, + "num_input_tokens_seen": 1650816, + "step": 1275 + }, + { + "epoch": 0.06254122590574841, + "grad_norm": 0.006774052977561951, + "learning_rate": 0.29924383444548974, + "loss": 0.1235, + "num_input_tokens_seen": 1657152, + "step": 1280 + }, + { + "epoch": 0.06278552756944275, + "grad_norm": 0.009479713626205921, + "learning_rate": 0.299237915748052, + "loss": 0.1398, + "num_input_tokens_seen": 1663360, + "step": 1285 + }, + { + "epoch": 0.06302982923313707, + "grad_norm": 0.008116134442389011, + "learning_rate": 0.2992319740362522, + "loss": 0.1351, + "num_input_tokens_seen": 1670080, + "step": 1290 + }, + { + "epoch": 0.06327413089683141, + "grad_norm": 0.005500311963260174, + "learning_rate": 0.2992260093110066, + "loss": 0.1341, + "num_input_tokens_seen": 1676992, + "step": 1295 + }, + { + "epoch": 0.06351843256052574, + "grad_norm": 0.008686036802828312, + "learning_rate": 0.2992200215732352, + "loss": 0.1567, + "num_input_tokens_seen": 1683168, + "step": 1300 + }, + { + "epoch": 0.06376273422422006, + "grad_norm": 0.008685456588864326, + "learning_rate": 0.2992140108238611, + "loss": 0.1587, + "num_input_tokens_seen": 1689632, + "step": 1305 + }, + { + "epoch": 0.0640070358879144, + "grad_norm": 0.010876053012907505, + "learning_rate": 0.2992079770638115, + "loss": 0.1663, + "num_input_tokens_seen": 1696320, + "step": 1310 + }, + { + "epoch": 0.06425133755160872, + "grad_norm": 0.012214096263051033, + "learning_rate": 0.29920192029401677, + "loss": 0.14, + "num_input_tokens_seen": 1702656, + "step": 1315 + }, + { + "epoch": 0.06449563921530306, + "grad_norm": 0.007783563807606697, + "learning_rate": 0.2991958405154109, + "loss": 0.1387, + "num_input_tokens_seen": 1709344, + "step": 1320 + }, + { + "epoch": 0.06473994087899738, + "grad_norm": 0.01277219783514738, + "learning_rate": 0.29918973772893154, + "loss": 0.1403, + "num_input_tokens_seen": 1716064, + "step": 1325 + }, + { + "epoch": 0.06498424254269172, + "grad_norm": 0.0073556979186832905, + "learning_rate": 0.29918361193551973, + "loss": 0.1286, + "num_input_tokens_seen": 1722592, + "step": 1330 + }, + { + "epoch": 0.06522854420638605, + "grad_norm": 0.00642261141911149, + "learning_rate": 0.29917746313612026, + "loss": 0.1553, + "num_input_tokens_seen": 1729120, + "step": 1335 + }, + { + "epoch": 0.06547284587008037, + "grad_norm": 0.01127227209508419, + "learning_rate": 0.29917129133168124, + "loss": 0.1401, + "num_input_tokens_seen": 1735424, + "step": 1340 + }, + { + "epoch": 0.06571714753377471, + "grad_norm": 0.010692550800740719, + "learning_rate": 0.2991650965231546, + "loss": 0.1394, + "num_input_tokens_seen": 1742208, + "step": 1345 + }, + { + "epoch": 0.06596144919746903, + "grad_norm": 0.01845361478626728, + "learning_rate": 0.29915887871149544, + "loss": 0.1241, + "num_input_tokens_seen": 1748736, + "step": 1350 + }, + { + "epoch": 0.06620575086116337, + "grad_norm": 0.0215369313955307, + "learning_rate": 0.2991526378976628, + "loss": 0.1597, + "num_input_tokens_seen": 1754912, + "step": 1355 + }, + { + "epoch": 0.0664500525248577, + "grad_norm": 0.01092858612537384, + "learning_rate": 0.29914637408261896, + "loss": 0.1534, + "num_input_tokens_seen": 1761120, + "step": 1360 + }, + { + "epoch": 0.06669435418855202, + "grad_norm": 0.014656784944236279, + "learning_rate": 0.29914008726733, + "loss": 0.1501, + "num_input_tokens_seen": 1767360, + "step": 1365 + }, + { + "epoch": 0.06693865585224636, + "grad_norm": 0.007447266019880772, + "learning_rate": 0.2991337774527653, + "loss": 0.1509, + "num_input_tokens_seen": 1774656, + "step": 1370 + }, + { + "epoch": 0.06718295751594068, + "grad_norm": 0.006443744990974665, + "learning_rate": 0.2991274446398981, + "loss": 0.1437, + "num_input_tokens_seen": 1781216, + "step": 1375 + }, + { + "epoch": 0.06742725917963502, + "grad_norm": 0.009226895868778229, + "learning_rate": 0.29912108882970484, + "loss": 0.1461, + "num_input_tokens_seen": 1788064, + "step": 1380 + }, + { + "epoch": 0.06767156084332934, + "grad_norm": 0.009342392906546593, + "learning_rate": 0.2991147100231657, + "loss": 0.1322, + "num_input_tokens_seen": 1794720, + "step": 1385 + }, + { + "epoch": 0.06791586250702367, + "grad_norm": 0.013391564600169659, + "learning_rate": 0.2991083082212644, + "loss": 0.1747, + "num_input_tokens_seen": 1800864, + "step": 1390 + }, + { + "epoch": 0.068160164170718, + "grad_norm": 0.006960629019886255, + "learning_rate": 0.2991018834249881, + "loss": 0.1532, + "num_input_tokens_seen": 1807424, + "step": 1395 + }, + { + "epoch": 0.06840446583441233, + "grad_norm": 0.007054580375552177, + "learning_rate": 0.29909543563532764, + "loss": 0.1425, + "num_input_tokens_seen": 1813856, + "step": 1400 + }, + { + "epoch": 0.06840446583441233, + "eval_loss": 0.15628843009471893, + "eval_runtime": 401.6972, + "eval_samples_per_second": 90.578, + "eval_steps_per_second": 22.646, + "num_input_tokens_seen": 1813856, + "step": 1400 + }, + { + "epoch": 0.06864876749810667, + "grad_norm": 0.006129363086074591, + "learning_rate": 0.29908896485327746, + "loss": 0.1628, + "num_input_tokens_seen": 1820544, + "step": 1405 + }, + { + "epoch": 0.06889306916180099, + "grad_norm": 0.010298346169292927, + "learning_rate": 0.29908247107983527, + "loss": 0.1355, + "num_input_tokens_seen": 1827616, + "step": 1410 + }, + { + "epoch": 0.06913737082549533, + "grad_norm": 0.008159324526786804, + "learning_rate": 0.29907595431600253, + "loss": 0.1442, + "num_input_tokens_seen": 1834112, + "step": 1415 + }, + { + "epoch": 0.06938167248918965, + "grad_norm": 0.006938764359802008, + "learning_rate": 0.29906941456278424, + "loss": 0.1853, + "num_input_tokens_seen": 1840320, + "step": 1420 + }, + { + "epoch": 0.06962597415288398, + "grad_norm": 0.008189276792109013, + "learning_rate": 0.2990628518211889, + "loss": 0.1556, + "num_input_tokens_seen": 1846816, + "step": 1425 + }, + { + "epoch": 0.06987027581657831, + "grad_norm": 0.006351861171424389, + "learning_rate": 0.2990562660922286, + "loss": 0.1426, + "num_input_tokens_seen": 1852864, + "step": 1430 + }, + { + "epoch": 0.07011457748027264, + "grad_norm": 0.008740276098251343, + "learning_rate": 0.2990496573769189, + "loss": 0.1626, + "num_input_tokens_seen": 1859328, + "step": 1435 + }, + { + "epoch": 0.07035887914396698, + "grad_norm": 0.011242554523050785, + "learning_rate": 0.29904302567627894, + "loss": 0.1585, + "num_input_tokens_seen": 1865856, + "step": 1440 + }, + { + "epoch": 0.0706031808076613, + "grad_norm": 0.005354073829948902, + "learning_rate": 0.2990363709913314, + "loss": 0.1412, + "num_input_tokens_seen": 1872672, + "step": 1445 + }, + { + "epoch": 0.07084748247135562, + "grad_norm": 0.010286780074238777, + "learning_rate": 0.29902969332310264, + "loss": 0.1372, + "num_input_tokens_seen": 1879392, + "step": 1450 + }, + { + "epoch": 0.07109178413504996, + "grad_norm": 0.007437588647007942, + "learning_rate": 0.2990229926726223, + "loss": 0.1409, + "num_input_tokens_seen": 1886688, + "step": 1455 + }, + { + "epoch": 0.07133608579874429, + "grad_norm": 0.005696033593267202, + "learning_rate": 0.29901626904092365, + "loss": 0.141, + "num_input_tokens_seen": 1892352, + "step": 1460 + }, + { + "epoch": 0.07158038746243862, + "grad_norm": 0.007549292407929897, + "learning_rate": 0.2990095224290438, + "loss": 0.131, + "num_input_tokens_seen": 1899136, + "step": 1465 + }, + { + "epoch": 0.07182468912613295, + "grad_norm": 0.011076409369707108, + "learning_rate": 0.29900275283802297, + "loss": 0.1452, + "num_input_tokens_seen": 1905856, + "step": 1470 + }, + { + "epoch": 0.07206899078982727, + "grad_norm": 0.00911741703748703, + "learning_rate": 0.2989959602689051, + "loss": 0.1368, + "num_input_tokens_seen": 1912608, + "step": 1475 + }, + { + "epoch": 0.07231329245352161, + "grad_norm": 0.0066538541577756405, + "learning_rate": 0.2989891447227379, + "loss": 0.1394, + "num_input_tokens_seen": 1918816, + "step": 1480 + }, + { + "epoch": 0.07255759411721593, + "grad_norm": 0.00867495033890009, + "learning_rate": 0.29898230620057215, + "loss": 0.1551, + "num_input_tokens_seen": 1925440, + "step": 1485 + }, + { + "epoch": 0.07280189578091027, + "grad_norm": 0.008712047711014748, + "learning_rate": 0.2989754447034626, + "loss": 0.1576, + "num_input_tokens_seen": 1932064, + "step": 1490 + }, + { + "epoch": 0.0730461974446046, + "grad_norm": 0.008222202770411968, + "learning_rate": 0.2989685602324673, + "loss": 0.1375, + "num_input_tokens_seen": 1938176, + "step": 1495 + }, + { + "epoch": 0.07329049910829893, + "grad_norm": 0.007401471026241779, + "learning_rate": 0.298961652788648, + "loss": 0.1486, + "num_input_tokens_seen": 1945152, + "step": 1500 + }, + { + "epoch": 0.07353480077199326, + "grad_norm": 0.011732184328138828, + "learning_rate": 0.29895472237306986, + "loss": 0.1612, + "num_input_tokens_seen": 1951520, + "step": 1505 + }, + { + "epoch": 0.07377910243568758, + "grad_norm": 0.009850456379354, + "learning_rate": 0.29894776898680164, + "loss": 0.1499, + "num_input_tokens_seen": 1958176, + "step": 1510 + }, + { + "epoch": 0.07402340409938192, + "grad_norm": 0.004411152098327875, + "learning_rate": 0.29894079263091566, + "loss": 0.1414, + "num_input_tokens_seen": 1964608, + "step": 1515 + }, + { + "epoch": 0.07426770576307624, + "grad_norm": 0.007934018969535828, + "learning_rate": 0.2989337933064877, + "loss": 0.1554, + "num_input_tokens_seen": 1971168, + "step": 1520 + }, + { + "epoch": 0.07451200742677058, + "grad_norm": 0.004671796690672636, + "learning_rate": 0.29892677101459725, + "loss": 0.1353, + "num_input_tokens_seen": 1977632, + "step": 1525 + }, + { + "epoch": 0.0747563090904649, + "grad_norm": 0.00719686783850193, + "learning_rate": 0.2989197257563272, + "loss": 0.1501, + "num_input_tokens_seen": 1984032, + "step": 1530 + }, + { + "epoch": 0.07500061075415923, + "grad_norm": 0.009209073148667812, + "learning_rate": 0.2989126575327639, + "loss": 0.1404, + "num_input_tokens_seen": 1990432, + "step": 1535 + }, + { + "epoch": 0.07524491241785357, + "grad_norm": 0.008410223759710789, + "learning_rate": 0.29890556634499754, + "loss": 0.1785, + "num_input_tokens_seen": 1996992, + "step": 1540 + }, + { + "epoch": 0.07548921408154789, + "grad_norm": 0.00932359229773283, + "learning_rate": 0.2988984521941216, + "loss": 0.1458, + "num_input_tokens_seen": 2004032, + "step": 1545 + }, + { + "epoch": 0.07573351574524223, + "grad_norm": 0.012650762684643269, + "learning_rate": 0.29889131508123307, + "loss": 0.1281, + "num_input_tokens_seen": 2010464, + "step": 1550 + }, + { + "epoch": 0.07597781740893655, + "grad_norm": 0.011608884669840336, + "learning_rate": 0.2988841550074327, + "loss": 0.1381, + "num_input_tokens_seen": 2016864, + "step": 1555 + }, + { + "epoch": 0.07622211907263088, + "grad_norm": 0.008648566901683807, + "learning_rate": 0.2988769719738246, + "loss": 0.1317, + "num_input_tokens_seen": 2023680, + "step": 1560 + }, + { + "epoch": 0.07646642073632522, + "grad_norm": 0.011406335979700089, + "learning_rate": 0.29886976598151666, + "loss": 0.1426, + "num_input_tokens_seen": 2030112, + "step": 1565 + }, + { + "epoch": 0.07671072240001954, + "grad_norm": 0.011077233590185642, + "learning_rate": 0.29886253703161986, + "loss": 0.1507, + "num_input_tokens_seen": 2036256, + "step": 1570 + }, + { + "epoch": 0.07695502406371388, + "grad_norm": 0.007399317342787981, + "learning_rate": 0.29885528512524917, + "loss": 0.1557, + "num_input_tokens_seen": 2042368, + "step": 1575 + }, + { + "epoch": 0.0771993257274082, + "grad_norm": 0.00637331185862422, + "learning_rate": 0.29884801026352287, + "loss": 0.1435, + "num_input_tokens_seen": 2048736, + "step": 1580 + }, + { + "epoch": 0.07744362739110254, + "grad_norm": 0.006028539966791868, + "learning_rate": 0.2988407124475629, + "loss": 0.1525, + "num_input_tokens_seen": 2055168, + "step": 1585 + }, + { + "epoch": 0.07768792905479686, + "grad_norm": 0.005999557673931122, + "learning_rate": 0.2988333916784945, + "loss": 0.1659, + "num_input_tokens_seen": 2061824, + "step": 1590 + }, + { + "epoch": 0.07793223071849119, + "grad_norm": 0.00962863303720951, + "learning_rate": 0.2988260479574468, + "loss": 0.1673, + "num_input_tokens_seen": 2068192, + "step": 1595 + }, + { + "epoch": 0.07817653238218553, + "grad_norm": 0.006079608108848333, + "learning_rate": 0.2988186812855523, + "loss": 0.1449, + "num_input_tokens_seen": 2074816, + "step": 1600 + }, + { + "epoch": 0.07817653238218553, + "eval_loss": 0.14754033088684082, + "eval_runtime": 402.0804, + "eval_samples_per_second": 90.492, + "eval_steps_per_second": 22.625, + "num_input_tokens_seen": 2074816, + "step": 1600 + }, + { + "epoch": 0.07842083404587985, + "grad_norm": 0.007888911291956902, + "learning_rate": 0.29881129166394693, + "loss": 0.1625, + "num_input_tokens_seen": 2081152, + "step": 1605 + }, + { + "epoch": 0.07866513570957419, + "grad_norm": 0.006740351673215628, + "learning_rate": 0.29880387909377026, + "loss": 0.1532, + "num_input_tokens_seen": 2087680, + "step": 1610 + }, + { + "epoch": 0.07890943737326851, + "grad_norm": 0.0048380764201283455, + "learning_rate": 0.2987964435761655, + "loss": 0.152, + "num_input_tokens_seen": 2094272, + "step": 1615 + }, + { + "epoch": 0.07915373903696284, + "grad_norm": 0.006759235635399818, + "learning_rate": 0.29878898511227925, + "loss": 0.1553, + "num_input_tokens_seen": 2100576, + "step": 1620 + }, + { + "epoch": 0.07939804070065717, + "grad_norm": 0.006417694501578808, + "learning_rate": 0.2987815037032617, + "loss": 0.1569, + "num_input_tokens_seen": 2106816, + "step": 1625 + }, + { + "epoch": 0.0796423423643515, + "grad_norm": 0.006444934289902449, + "learning_rate": 0.29877399935026655, + "loss": 0.1604, + "num_input_tokens_seen": 2112960, + "step": 1630 + }, + { + "epoch": 0.07988664402804584, + "grad_norm": 0.008458716794848442, + "learning_rate": 0.2987664720544511, + "loss": 0.1424, + "num_input_tokens_seen": 2119360, + "step": 1635 + }, + { + "epoch": 0.08013094569174016, + "grad_norm": 0.005090838763862848, + "learning_rate": 0.2987589218169761, + "loss": 0.1333, + "num_input_tokens_seen": 2126048, + "step": 1640 + }, + { + "epoch": 0.08037524735543448, + "grad_norm": 0.00973370112478733, + "learning_rate": 0.29875134863900604, + "loss": 0.1248, + "num_input_tokens_seen": 2132736, + "step": 1645 + }, + { + "epoch": 0.08061954901912882, + "grad_norm": 0.007689916528761387, + "learning_rate": 0.29874375252170865, + "loss": 0.1184, + "num_input_tokens_seen": 2138880, + "step": 1650 + }, + { + "epoch": 0.08086385068282315, + "grad_norm": 0.016372185200452805, + "learning_rate": 0.2987361334662553, + "loss": 0.1192, + "num_input_tokens_seen": 2145856, + "step": 1655 + }, + { + "epoch": 0.08110815234651748, + "grad_norm": 0.008520865812897682, + "learning_rate": 0.29872849147382113, + "loss": 0.1245, + "num_input_tokens_seen": 2152480, + "step": 1660 + }, + { + "epoch": 0.08135245401021181, + "grad_norm": 0.011617694981396198, + "learning_rate": 0.2987208265455845, + "loss": 0.1367, + "num_input_tokens_seen": 2158720, + "step": 1665 + }, + { + "epoch": 0.08159675567390615, + "grad_norm": 0.006529188249260187, + "learning_rate": 0.29871313868272753, + "loss": 0.1569, + "num_input_tokens_seen": 2165536, + "step": 1670 + }, + { + "epoch": 0.08184105733760047, + "grad_norm": 0.00959371030330658, + "learning_rate": 0.29870542788643567, + "loss": 0.1626, + "num_input_tokens_seen": 2172128, + "step": 1675 + }, + { + "epoch": 0.0820853590012948, + "grad_norm": 0.004780045244842768, + "learning_rate": 0.2986976941578981, + "loss": 0.1405, + "num_input_tokens_seen": 2178400, + "step": 1680 + }, + { + "epoch": 0.08232966066498913, + "grad_norm": 0.004448923747986555, + "learning_rate": 0.29868993749830747, + "loss": 0.1553, + "num_input_tokens_seen": 2184736, + "step": 1685 + }, + { + "epoch": 0.08257396232868346, + "grad_norm": 0.009723754599690437, + "learning_rate": 0.2986821579088598, + "loss": 0.135, + "num_input_tokens_seen": 2191328, + "step": 1690 + }, + { + "epoch": 0.0828182639923778, + "grad_norm": 0.006486440543085337, + "learning_rate": 0.29867435539075504, + "loss": 0.1593, + "num_input_tokens_seen": 2197632, + "step": 1695 + }, + { + "epoch": 0.08306256565607212, + "grad_norm": 0.007278315722942352, + "learning_rate": 0.2986665299451963, + "loss": 0.1443, + "num_input_tokens_seen": 2204000, + "step": 1700 + }, + { + "epoch": 0.08330686731976644, + "grad_norm": 0.006257487460970879, + "learning_rate": 0.29865868157339037, + "loss": 0.1534, + "num_input_tokens_seen": 2210272, + "step": 1705 + }, + { + "epoch": 0.08355116898346078, + "grad_norm": 0.009069254621863365, + "learning_rate": 0.2986508102765476, + "loss": 0.1159, + "num_input_tokens_seen": 2217248, + "step": 1710 + }, + { + "epoch": 0.0837954706471551, + "grad_norm": 0.011428210884332657, + "learning_rate": 0.2986429160558818, + "loss": 0.1305, + "num_input_tokens_seen": 2223584, + "step": 1715 + }, + { + "epoch": 0.08403977231084944, + "grad_norm": 0.010025345720350742, + "learning_rate": 0.2986349989126104, + "loss": 0.1456, + "num_input_tokens_seen": 2230720, + "step": 1720 + }, + { + "epoch": 0.08428407397454377, + "grad_norm": 0.011537850834429264, + "learning_rate": 0.29862705884795426, + "loss": 0.1672, + "num_input_tokens_seen": 2237216, + "step": 1725 + }, + { + "epoch": 0.08452837563823809, + "grad_norm": 0.007283071056008339, + "learning_rate": 0.2986190958631379, + "loss": 0.1424, + "num_input_tokens_seen": 2243584, + "step": 1730 + }, + { + "epoch": 0.08477267730193243, + "grad_norm": 0.008740660734474659, + "learning_rate": 0.29861110995938933, + "loss": 0.14, + "num_input_tokens_seen": 2250240, + "step": 1735 + }, + { + "epoch": 0.08501697896562675, + "grad_norm": 0.008466245606541634, + "learning_rate": 0.29860310113794, + "loss": 0.1751, + "num_input_tokens_seen": 2256480, + "step": 1740 + }, + { + "epoch": 0.08526128062932109, + "grad_norm": 0.017457470297813416, + "learning_rate": 0.29859506940002506, + "loss": 0.1472, + "num_input_tokens_seen": 2263232, + "step": 1745 + }, + { + "epoch": 0.08550558229301541, + "grad_norm": 0.00695843854919076, + "learning_rate": 0.298587014746883, + "loss": 0.1509, + "num_input_tokens_seen": 2269568, + "step": 1750 + }, + { + "epoch": 0.08574988395670975, + "grad_norm": 0.004921743646264076, + "learning_rate": 0.298578937179756, + "loss": 0.1317, + "num_input_tokens_seen": 2275840, + "step": 1755 + }, + { + "epoch": 0.08599418562040408, + "grad_norm": 0.011255567893385887, + "learning_rate": 0.29857083669988976, + "loss": 0.1598, + "num_input_tokens_seen": 2281824, + "step": 1760 + }, + { + "epoch": 0.0862384872840984, + "grad_norm": 0.006461360491812229, + "learning_rate": 0.29856271330853346, + "loss": 0.114, + "num_input_tokens_seen": 2288288, + "step": 1765 + }, + { + "epoch": 0.08648278894779274, + "grad_norm": 0.005772430915385485, + "learning_rate": 0.2985545670069398, + "loss": 0.156, + "num_input_tokens_seen": 2294496, + "step": 1770 + }, + { + "epoch": 0.08672709061148706, + "grad_norm": 0.009369258768856525, + "learning_rate": 0.29854639779636505, + "loss": 0.1556, + "num_input_tokens_seen": 2300992, + "step": 1775 + }, + { + "epoch": 0.0869713922751814, + "grad_norm": 0.009414736181497574, + "learning_rate": 0.298538205678069, + "loss": 0.1505, + "num_input_tokens_seen": 2307264, + "step": 1780 + }, + { + "epoch": 0.08721569393887572, + "grad_norm": 0.0056551494635641575, + "learning_rate": 0.298529990653315, + "loss": 0.1382, + "num_input_tokens_seen": 2313568, + "step": 1785 + }, + { + "epoch": 0.08745999560257005, + "grad_norm": 0.007361214607954025, + "learning_rate": 0.29852175272336984, + "loss": 0.1401, + "num_input_tokens_seen": 2319808, + "step": 1790 + }, + { + "epoch": 0.08770429726626439, + "grad_norm": 0.010200964286923409, + "learning_rate": 0.29851349188950405, + "loss": 0.1573, + "num_input_tokens_seen": 2326208, + "step": 1795 + }, + { + "epoch": 0.08794859892995871, + "grad_norm": 0.007126512471586466, + "learning_rate": 0.2985052081529914, + "loss": 0.1475, + "num_input_tokens_seen": 2332544, + "step": 1800 + }, + { + "epoch": 0.08794859892995871, + "eval_loss": 0.158894345164299, + "eval_runtime": 401.876, + "eval_samples_per_second": 90.538, + "eval_steps_per_second": 22.636, + "num_input_tokens_seen": 2332544, + "step": 1800 + }, + { + "epoch": 0.08819290059365305, + "grad_norm": 0.0073537821881473064, + "learning_rate": 0.29849690151510944, + "loss": 0.1904, + "num_input_tokens_seen": 2338784, + "step": 1805 + }, + { + "epoch": 0.08843720225734737, + "grad_norm": 0.00714768934994936, + "learning_rate": 0.2984885719771392, + "loss": 0.1597, + "num_input_tokens_seen": 2345152, + "step": 1810 + }, + { + "epoch": 0.0886815039210417, + "grad_norm": 0.0045191021636128426, + "learning_rate": 0.2984802195403651, + "loss": 0.1432, + "num_input_tokens_seen": 2351744, + "step": 1815 + }, + { + "epoch": 0.08892580558473603, + "grad_norm": 0.006616136524826288, + "learning_rate": 0.2984718442060752, + "loss": 0.1336, + "num_input_tokens_seen": 2358336, + "step": 1820 + }, + { + "epoch": 0.08917010724843036, + "grad_norm": 0.006773095577955246, + "learning_rate": 0.2984634459755611, + "loss": 0.1454, + "num_input_tokens_seen": 2365056, + "step": 1825 + }, + { + "epoch": 0.0894144089121247, + "grad_norm": 0.007107454352080822, + "learning_rate": 0.29845502485011793, + "loss": 0.1494, + "num_input_tokens_seen": 2371424, + "step": 1830 + }, + { + "epoch": 0.08965871057581902, + "grad_norm": 0.0095054404810071, + "learning_rate": 0.2984465808310444, + "loss": 0.1506, + "num_input_tokens_seen": 2378240, + "step": 1835 + }, + { + "epoch": 0.08990301223951336, + "grad_norm": 0.006321938708424568, + "learning_rate": 0.29843811391964253, + "loss": 0.14, + "num_input_tokens_seen": 2384832, + "step": 1840 + }, + { + "epoch": 0.09014731390320768, + "grad_norm": 0.010147669352591038, + "learning_rate": 0.2984296241172182, + "loss": 0.1479, + "num_input_tokens_seen": 2391648, + "step": 1845 + }, + { + "epoch": 0.090391615566902, + "grad_norm": 0.0045564137399196625, + "learning_rate": 0.29842111142508043, + "loss": 0.1315, + "num_input_tokens_seen": 2397856, + "step": 1850 + }, + { + "epoch": 0.09063591723059634, + "grad_norm": 0.009637849405407906, + "learning_rate": 0.29841257584454217, + "loss": 0.1578, + "num_input_tokens_seen": 2404096, + "step": 1855 + }, + { + "epoch": 0.09088021889429067, + "grad_norm": 0.008207933977246284, + "learning_rate": 0.29840401737691963, + "loss": 0.1621, + "num_input_tokens_seen": 2411008, + "step": 1860 + }, + { + "epoch": 0.091124520557985, + "grad_norm": 0.016169443726539612, + "learning_rate": 0.29839543602353263, + "loss": 0.1711, + "num_input_tokens_seen": 2417376, + "step": 1865 + }, + { + "epoch": 0.09136882222167933, + "grad_norm": 0.006187734194099903, + "learning_rate": 0.2983868317857046, + "loss": 0.1606, + "num_input_tokens_seen": 2423776, + "step": 1870 + }, + { + "epoch": 0.09161312388537365, + "grad_norm": 0.005456337705254555, + "learning_rate": 0.2983782046647623, + "loss": 0.1624, + "num_input_tokens_seen": 2430560, + "step": 1875 + }, + { + "epoch": 0.09185742554906799, + "grad_norm": 0.0069424365647137165, + "learning_rate": 0.2983695546620362, + "loss": 0.1442, + "num_input_tokens_seen": 2436864, + "step": 1880 + }, + { + "epoch": 0.09210172721276232, + "grad_norm": 0.005672098137438297, + "learning_rate": 0.2983608817788603, + "loss": 0.1482, + "num_input_tokens_seen": 2443136, + "step": 1885 + }, + { + "epoch": 0.09234602887645665, + "grad_norm": 0.00938822329044342, + "learning_rate": 0.29835218601657193, + "loss": 0.1476, + "num_input_tokens_seen": 2449824, + "step": 1890 + }, + { + "epoch": 0.09259033054015098, + "grad_norm": 0.010361879132688046, + "learning_rate": 0.29834346737651224, + "loss": 0.1388, + "num_input_tokens_seen": 2456704, + "step": 1895 + }, + { + "epoch": 0.0928346322038453, + "grad_norm": 0.006879544351249933, + "learning_rate": 0.29833472586002563, + "loss": 0.145, + "num_input_tokens_seen": 2463360, + "step": 1900 + }, + { + "epoch": 0.09307893386753964, + "grad_norm": 0.007334616966545582, + "learning_rate": 0.29832596146846024, + "loss": 0.1397, + "num_input_tokens_seen": 2470112, + "step": 1905 + }, + { + "epoch": 0.09332323553123396, + "grad_norm": 0.013973837718367577, + "learning_rate": 0.2983171742031676, + "loss": 0.1641, + "num_input_tokens_seen": 2476256, + "step": 1910 + }, + { + "epoch": 0.0935675371949283, + "grad_norm": 0.006392891984432936, + "learning_rate": 0.2983083640655028, + "loss": 0.1689, + "num_input_tokens_seen": 2482336, + "step": 1915 + }, + { + "epoch": 0.09381183885862263, + "grad_norm": 0.005648659076541662, + "learning_rate": 0.29829953105682455, + "loss": 0.1539, + "num_input_tokens_seen": 2488320, + "step": 1920 + }, + { + "epoch": 0.09405614052231696, + "grad_norm": 0.006061872001737356, + "learning_rate": 0.29829067517849495, + "loss": 0.1491, + "num_input_tokens_seen": 2494944, + "step": 1925 + }, + { + "epoch": 0.09430044218601129, + "grad_norm": 0.008424672298133373, + "learning_rate": 0.2982817964318797, + "loss": 0.1266, + "num_input_tokens_seen": 2501568, + "step": 1930 + }, + { + "epoch": 0.09454474384970561, + "grad_norm": 0.004992188885807991, + "learning_rate": 0.298272894818348, + "loss": 0.1392, + "num_input_tokens_seen": 2508224, + "step": 1935 + }, + { + "epoch": 0.09478904551339995, + "grad_norm": 0.011382006108760834, + "learning_rate": 0.2982639703392726, + "loss": 0.1675, + "num_input_tokens_seen": 2515360, + "step": 1940 + }, + { + "epoch": 0.09503334717709427, + "grad_norm": 0.00839060265570879, + "learning_rate": 0.29825502299602974, + "loss": 0.1398, + "num_input_tokens_seen": 2522080, + "step": 1945 + }, + { + "epoch": 0.09527764884078861, + "grad_norm": 0.005447382107377052, + "learning_rate": 0.2982460527899993, + "loss": 0.1223, + "num_input_tokens_seen": 2528736, + "step": 1950 + }, + { + "epoch": 0.09552195050448294, + "grad_norm": 0.009465738199651241, + "learning_rate": 0.29823705972256453, + "loss": 0.1639, + "num_input_tokens_seen": 2534880, + "step": 1955 + }, + { + "epoch": 0.09576625216817726, + "grad_norm": 0.009535813704133034, + "learning_rate": 0.2982280437951123, + "loss": 0.1606, + "num_input_tokens_seen": 2541472, + "step": 1960 + }, + { + "epoch": 0.0960105538318716, + "grad_norm": 0.0053486330434679985, + "learning_rate": 0.298219005009033, + "loss": 0.1329, + "num_input_tokens_seen": 2548320, + "step": 1965 + }, + { + "epoch": 0.09625485549556592, + "grad_norm": 0.008033198304474354, + "learning_rate": 0.29820994336572043, + "loss": 0.1331, + "num_input_tokens_seen": 2554944, + "step": 1970 + }, + { + "epoch": 0.09649915715926026, + "grad_norm": 0.007797688245773315, + "learning_rate": 0.2982008588665721, + "loss": 0.1637, + "num_input_tokens_seen": 2561056, + "step": 1975 + }, + { + "epoch": 0.09674345882295458, + "grad_norm": 0.007322000339627266, + "learning_rate": 0.2981917515129889, + "loss": 0.1313, + "num_input_tokens_seen": 2567488, + "step": 1980 + }, + { + "epoch": 0.09698776048664891, + "grad_norm": 0.009402529336512089, + "learning_rate": 0.2981826213063753, + "loss": 0.1637, + "num_input_tokens_seen": 2574176, + "step": 1985 + }, + { + "epoch": 0.09723206215034325, + "grad_norm": 0.0062019540928304195, + "learning_rate": 0.2981734682481394, + "loss": 0.1421, + "num_input_tokens_seen": 2580992, + "step": 1990 + }, + { + "epoch": 0.09747636381403757, + "grad_norm": 0.006288324482738972, + "learning_rate": 0.29816429233969255, + "loss": 0.1662, + "num_input_tokens_seen": 2587712, + "step": 1995 + }, + { + "epoch": 0.09772066547773191, + "grad_norm": 0.006934609264135361, + "learning_rate": 0.2981550935824499, + "loss": 0.137, + "num_input_tokens_seen": 2594720, + "step": 2000 + }, + { + "epoch": 0.09772066547773191, + "eval_loss": 0.14684335887432098, + "eval_runtime": 402.4778, + "eval_samples_per_second": 90.403, + "eval_steps_per_second": 22.602, + "num_input_tokens_seen": 2594720, + "step": 2000 + }, + { + "epoch": 0.09796496714142623, + "grad_norm": 0.009815394878387451, + "learning_rate": 0.29814587197783, + "loss": 0.1491, + "num_input_tokens_seen": 2600928, + "step": 2005 + }, + { + "epoch": 0.09820926880512057, + "grad_norm": 0.006815847009420395, + "learning_rate": 0.29813662752725495, + "loss": 0.1439, + "num_input_tokens_seen": 2607488, + "step": 2010 + }, + { + "epoch": 0.0984535704688149, + "grad_norm": 0.005136177409440279, + "learning_rate": 0.29812736023215025, + "loss": 0.1423, + "num_input_tokens_seen": 2613984, + "step": 2015 + }, + { + "epoch": 0.09869787213250922, + "grad_norm": 0.006002115551382303, + "learning_rate": 0.29811807009394514, + "loss": 0.1212, + "num_input_tokens_seen": 2620832, + "step": 2020 + }, + { + "epoch": 0.09894217379620356, + "grad_norm": 0.008059307001531124, + "learning_rate": 0.2981087571140723, + "loss": 0.1583, + "num_input_tokens_seen": 2627104, + "step": 2025 + }, + { + "epoch": 0.09918647545989788, + "grad_norm": 0.0053268130868673325, + "learning_rate": 0.2980994212939678, + "loss": 0.1334, + "num_input_tokens_seen": 2633504, + "step": 2030 + }, + { + "epoch": 0.09943077712359222, + "grad_norm": 0.010707980953156948, + "learning_rate": 0.2980900626350715, + "loss": 0.1296, + "num_input_tokens_seen": 2639648, + "step": 2035 + }, + { + "epoch": 0.09967507878728654, + "grad_norm": 0.00944272056221962, + "learning_rate": 0.29808068113882646, + "loss": 0.1275, + "num_input_tokens_seen": 2645984, + "step": 2040 + }, + { + "epoch": 0.09991938045098087, + "grad_norm": 0.0067657423205673695, + "learning_rate": 0.2980712768066795, + "loss": 0.1216, + "num_input_tokens_seen": 2652576, + "step": 2045 + }, + { + "epoch": 0.1001636821146752, + "grad_norm": 0.007116362918168306, + "learning_rate": 0.2980618496400809, + "loss": 0.1644, + "num_input_tokens_seen": 2659008, + "step": 2050 + }, + { + "epoch": 0.10040798377836953, + "grad_norm": 0.011232172138988972, + "learning_rate": 0.2980523996404844, + "loss": 0.176, + "num_input_tokens_seen": 2665280, + "step": 2055 + }, + { + "epoch": 0.10065228544206387, + "grad_norm": 0.00566619448363781, + "learning_rate": 0.2980429268093473, + "loss": 0.1458, + "num_input_tokens_seen": 2671584, + "step": 2060 + }, + { + "epoch": 0.10089658710575819, + "grad_norm": 0.004092557355761528, + "learning_rate": 0.29803343114813047, + "loss": 0.1437, + "num_input_tokens_seen": 2677792, + "step": 2065 + }, + { + "epoch": 0.10114088876945251, + "grad_norm": 0.0036944961175322533, + "learning_rate": 0.2980239126582983, + "loss": 0.1109, + "num_input_tokens_seen": 2684192, + "step": 2070 + }, + { + "epoch": 0.10138519043314685, + "grad_norm": 0.007944513112306595, + "learning_rate": 0.2980143713413186, + "loss": 0.1519, + "num_input_tokens_seen": 2690624, + "step": 2075 + }, + { + "epoch": 0.10162949209684118, + "grad_norm": 0.011847447603940964, + "learning_rate": 0.29800480719866274, + "loss": 0.1553, + "num_input_tokens_seen": 2697088, + "step": 2080 + }, + { + "epoch": 0.10187379376053551, + "grad_norm": 0.009012926369905472, + "learning_rate": 0.2979952202318057, + "loss": 0.1532, + "num_input_tokens_seen": 2703232, + "step": 2085 + }, + { + "epoch": 0.10211809542422984, + "grad_norm": 0.00517695490270853, + "learning_rate": 0.2979856104422259, + "loss": 0.1369, + "num_input_tokens_seen": 2709824, + "step": 2090 + }, + { + "epoch": 0.10236239708792418, + "grad_norm": 0.010814575478434563, + "learning_rate": 0.2979759778314052, + "loss": 0.1432, + "num_input_tokens_seen": 2716032, + "step": 2095 + }, + { + "epoch": 0.1026066987516185, + "grad_norm": 0.006288168951869011, + "learning_rate": 0.2979663224008292, + "loss": 0.1528, + "num_input_tokens_seen": 2722048, + "step": 2100 + }, + { + "epoch": 0.10285100041531282, + "grad_norm": 0.011449499987065792, + "learning_rate": 0.2979566441519868, + "loss": 0.1385, + "num_input_tokens_seen": 2728416, + "step": 2105 + }, + { + "epoch": 0.10309530207900716, + "grad_norm": 0.005992741324007511, + "learning_rate": 0.29794694308637054, + "loss": 0.169, + "num_input_tokens_seen": 2734720, + "step": 2110 + }, + { + "epoch": 0.10333960374270149, + "grad_norm": 0.014546139165759087, + "learning_rate": 0.2979372192054764, + "loss": 0.1521, + "num_input_tokens_seen": 2741376, + "step": 2115 + }, + { + "epoch": 0.10358390540639582, + "grad_norm": 0.0036994419060647488, + "learning_rate": 0.297927472510804, + "loss": 0.131, + "num_input_tokens_seen": 2747552, + "step": 2120 + }, + { + "epoch": 0.10382820707009015, + "grad_norm": 0.006187394727021456, + "learning_rate": 0.29791770300385634, + "loss": 0.1525, + "num_input_tokens_seen": 2753824, + "step": 2125 + }, + { + "epoch": 0.10407250873378447, + "grad_norm": 0.006038791500031948, + "learning_rate": 0.29790791068614003, + "loss": 0.1518, + "num_input_tokens_seen": 2760256, + "step": 2130 + }, + { + "epoch": 0.10431681039747881, + "grad_norm": 0.006093179807066917, + "learning_rate": 0.2978980955591652, + "loss": 0.1482, + "num_input_tokens_seen": 2767264, + "step": 2135 + }, + { + "epoch": 0.10456111206117313, + "grad_norm": 0.004780122544616461, + "learning_rate": 0.2978882576244454, + "loss": 0.1204, + "num_input_tokens_seen": 2773920, + "step": 2140 + }, + { + "epoch": 0.10480541372486747, + "grad_norm": 0.004384675528854132, + "learning_rate": 0.2978783968834978, + "loss": 0.1251, + "num_input_tokens_seen": 2780352, + "step": 2145 + }, + { + "epoch": 0.1050497153885618, + "grad_norm": 0.006422858219593763, + "learning_rate": 0.29786851333784303, + "loss": 0.123, + "num_input_tokens_seen": 2787584, + "step": 2150 + }, + { + "epoch": 0.10529401705225612, + "grad_norm": 0.012434789910912514, + "learning_rate": 0.2978586069890053, + "loss": 0.1446, + "num_input_tokens_seen": 2793888, + "step": 2155 + }, + { + "epoch": 0.10553831871595046, + "grad_norm": 0.009952078573405743, + "learning_rate": 0.29784867783851227, + "loss": 0.1864, + "num_input_tokens_seen": 2800480, + "step": 2160 + }, + { + "epoch": 0.10578262037964478, + "grad_norm": 0.006644707638770342, + "learning_rate": 0.2978387258878951, + "loss": 0.1646, + "num_input_tokens_seen": 2806944, + "step": 2165 + }, + { + "epoch": 0.10602692204333912, + "grad_norm": 0.005006655119359493, + "learning_rate": 0.29782875113868856, + "loss": 0.1253, + "num_input_tokens_seen": 2814176, + "step": 2170 + }, + { + "epoch": 0.10627122370703344, + "grad_norm": 0.009269512258470058, + "learning_rate": 0.2978187535924309, + "loss": 0.157, + "num_input_tokens_seen": 2820768, + "step": 2175 + }, + { + "epoch": 0.10651552537072777, + "grad_norm": 0.0166134275496006, + "learning_rate": 0.29780873325066376, + "loss": 0.1613, + "num_input_tokens_seen": 2827424, + "step": 2180 + }, + { + "epoch": 0.1067598270344221, + "grad_norm": 0.005827858578413725, + "learning_rate": 0.2977986901149325, + "loss": 0.1527, + "num_input_tokens_seen": 2833920, + "step": 2185 + }, + { + "epoch": 0.10700412869811643, + "grad_norm": 0.005412808153778315, + "learning_rate": 0.29778862418678587, + "loss": 0.1617, + "num_input_tokens_seen": 2840352, + "step": 2190 + }, + { + "epoch": 0.10724843036181077, + "grad_norm": 0.009247125126421452, + "learning_rate": 0.29777853546777616, + "loss": 0.1592, + "num_input_tokens_seen": 2847456, + "step": 2195 + }, + { + "epoch": 0.10749273202550509, + "grad_norm": 0.007482137996703386, + "learning_rate": 0.2977684239594592, + "loss": 0.1557, + "num_input_tokens_seen": 2853856, + "step": 2200 + }, + { + "epoch": 0.10749273202550509, + "eval_loss": 0.16597618162631989, + "eval_runtime": 401.9587, + "eval_samples_per_second": 90.519, + "eval_steps_per_second": 22.632, + "num_input_tokens_seen": 2853856, + "step": 2200 + }, + { + "epoch": 0.10773703368919943, + "grad_norm": 0.004710840526968241, + "learning_rate": 0.29775828966339424, + "loss": 0.1563, + "num_input_tokens_seen": 2860320, + "step": 2205 + }, + { + "epoch": 0.10798133535289375, + "grad_norm": 0.007008232641965151, + "learning_rate": 0.29774813258114424, + "loss": 0.1513, + "num_input_tokens_seen": 2866432, + "step": 2210 + }, + { + "epoch": 0.10822563701658808, + "grad_norm": 0.003906615544110537, + "learning_rate": 0.29773795271427544, + "loss": 0.131, + "num_input_tokens_seen": 2872864, + "step": 2215 + }, + { + "epoch": 0.10846993868028242, + "grad_norm": 0.006734485272318125, + "learning_rate": 0.2977277500643577, + "loss": 0.1497, + "num_input_tokens_seen": 2879296, + "step": 2220 + }, + { + "epoch": 0.10871424034397674, + "grad_norm": 0.005524495150893927, + "learning_rate": 0.29771752463296447, + "loss": 0.1467, + "num_input_tokens_seen": 2885344, + "step": 2225 + }, + { + "epoch": 0.10895854200767108, + "grad_norm": 0.00877354945987463, + "learning_rate": 0.29770727642167266, + "loss": 0.1392, + "num_input_tokens_seen": 2891744, + "step": 2230 + }, + { + "epoch": 0.1092028436713654, + "grad_norm": 0.011368148028850555, + "learning_rate": 0.29769700543206257, + "loss": 0.141, + "num_input_tokens_seen": 2898368, + "step": 2235 + }, + { + "epoch": 0.10944714533505973, + "grad_norm": 0.00859091803431511, + "learning_rate": 0.2976867116657182, + "loss": 0.1708, + "num_input_tokens_seen": 2904640, + "step": 2240 + }, + { + "epoch": 0.10969144699875406, + "grad_norm": 0.006714307703077793, + "learning_rate": 0.2976763951242269, + "loss": 0.1368, + "num_input_tokens_seen": 2910880, + "step": 2245 + }, + { + "epoch": 0.10993574866244839, + "grad_norm": 0.0072138430550694466, + "learning_rate": 0.29766605580917965, + "loss": 0.1582, + "num_input_tokens_seen": 2917376, + "step": 2250 + }, + { + "epoch": 0.11018005032614273, + "grad_norm": 0.005558820441365242, + "learning_rate": 0.29765569372217093, + "loss": 0.1497, + "num_input_tokens_seen": 2923936, + "step": 2255 + }, + { + "epoch": 0.11042435198983705, + "grad_norm": 0.0052739097736775875, + "learning_rate": 0.2976453088647987, + "loss": 0.1399, + "num_input_tokens_seen": 2930784, + "step": 2260 + }, + { + "epoch": 0.11066865365353137, + "grad_norm": 0.010841584764420986, + "learning_rate": 0.2976349012386644, + "loss": 0.1648, + "num_input_tokens_seen": 2937056, + "step": 2265 + }, + { + "epoch": 0.11091295531722571, + "grad_norm": 0.008292659185826778, + "learning_rate": 0.29762447084537297, + "loss": 0.1389, + "num_input_tokens_seen": 2943648, + "step": 2270 + }, + { + "epoch": 0.11115725698092004, + "grad_norm": 0.005531892646104097, + "learning_rate": 0.29761401768653306, + "loss": 0.1302, + "num_input_tokens_seen": 2950816, + "step": 2275 + }, + { + "epoch": 0.11140155864461437, + "grad_norm": 0.00612834095954895, + "learning_rate": 0.29760354176375653, + "loss": 0.1364, + "num_input_tokens_seen": 2957088, + "step": 2280 + }, + { + "epoch": 0.1116458603083087, + "grad_norm": 0.010711456649005413, + "learning_rate": 0.29759304307865897, + "loss": 0.1521, + "num_input_tokens_seen": 2963680, + "step": 2285 + }, + { + "epoch": 0.11189016197200304, + "grad_norm": 0.008454322814941406, + "learning_rate": 0.2975825216328594, + "loss": 0.1592, + "num_input_tokens_seen": 2970016, + "step": 2290 + }, + { + "epoch": 0.11213446363569736, + "grad_norm": 0.010183153674006462, + "learning_rate": 0.2975719774279804, + "loss": 0.1358, + "num_input_tokens_seen": 2976544, + "step": 2295 + }, + { + "epoch": 0.11237876529939168, + "grad_norm": 0.007332892622798681, + "learning_rate": 0.29756141046564794, + "loss": 0.136, + "num_input_tokens_seen": 2982688, + "step": 2300 + }, + { + "epoch": 0.11262306696308602, + "grad_norm": 0.015652349218726158, + "learning_rate": 0.2975508207474916, + "loss": 0.1639, + "num_input_tokens_seen": 2988960, + "step": 2305 + }, + { + "epoch": 0.11286736862678035, + "grad_norm": 0.01390936691313982, + "learning_rate": 0.2975402082751445, + "loss": 0.1442, + "num_input_tokens_seen": 2995328, + "step": 2310 + }, + { + "epoch": 0.11311167029047468, + "grad_norm": 0.004764043726027012, + "learning_rate": 0.29752957305024313, + "loss": 0.168, + "num_input_tokens_seen": 3002240, + "step": 2315 + }, + { + "epoch": 0.11335597195416901, + "grad_norm": 0.009020810946822166, + "learning_rate": 0.2975189150744277, + "loss": 0.1598, + "num_input_tokens_seen": 3009408, + "step": 2320 + }, + { + "epoch": 0.11360027361786333, + "grad_norm": 0.007300050463527441, + "learning_rate": 0.29750823434934165, + "loss": 0.1663, + "num_input_tokens_seen": 3015488, + "step": 2325 + }, + { + "epoch": 0.11384457528155767, + "grad_norm": 0.004080094397068024, + "learning_rate": 0.29749753087663217, + "loss": 0.1427, + "num_input_tokens_seen": 3022528, + "step": 2330 + }, + { + "epoch": 0.114088876945252, + "grad_norm": 0.005133760627359152, + "learning_rate": 0.29748680465794985, + "loss": 0.1172, + "num_input_tokens_seen": 3029088, + "step": 2335 + }, + { + "epoch": 0.11433317860894633, + "grad_norm": 0.006805312354117632, + "learning_rate": 0.29747605569494884, + "loss": 0.1638, + "num_input_tokens_seen": 3035648, + "step": 2340 + }, + { + "epoch": 0.11457748027264066, + "grad_norm": 0.009152105078101158, + "learning_rate": 0.29746528398928673, + "loss": 0.1494, + "num_input_tokens_seen": 3042016, + "step": 2345 + }, + { + "epoch": 0.11482178193633498, + "grad_norm": 0.005854794289916754, + "learning_rate": 0.2974544895426247, + "loss": 0.1353, + "num_input_tokens_seen": 3048928, + "step": 2350 + }, + { + "epoch": 0.11506608360002932, + "grad_norm": 0.005532880313694477, + "learning_rate": 0.29744367235662733, + "loss": 0.1459, + "num_input_tokens_seen": 3055168, + "step": 2355 + }, + { + "epoch": 0.11531038526372364, + "grad_norm": 0.005569893401116133, + "learning_rate": 0.29743283243296276, + "loss": 0.1541, + "num_input_tokens_seen": 3061120, + "step": 2360 + }, + { + "epoch": 0.11555468692741798, + "grad_norm": 0.0055623906664550304, + "learning_rate": 0.29742196977330276, + "loss": 0.1543, + "num_input_tokens_seen": 3067584, + "step": 2365 + }, + { + "epoch": 0.1157989885911123, + "grad_norm": 0.006891594268381596, + "learning_rate": 0.2974110843793223, + "loss": 0.1523, + "num_input_tokens_seen": 3073728, + "step": 2370 + }, + { + "epoch": 0.11604329025480664, + "grad_norm": 0.007641185075044632, + "learning_rate": 0.2974001762527002, + "loss": 0.1179, + "num_input_tokens_seen": 3080416, + "step": 2375 + }, + { + "epoch": 0.11628759191850097, + "grad_norm": 0.0062032416462898254, + "learning_rate": 0.2973892453951186, + "loss": 0.1598, + "num_input_tokens_seen": 3086912, + "step": 2380 + }, + { + "epoch": 0.11653189358219529, + "grad_norm": 0.007384798023849726, + "learning_rate": 0.2973782918082631, + "loss": 0.128, + "num_input_tokens_seen": 3093472, + "step": 2385 + }, + { + "epoch": 0.11677619524588963, + "grad_norm": 0.00857765693217516, + "learning_rate": 0.29736731549382295, + "loss": 0.1571, + "num_input_tokens_seen": 3100096, + "step": 2390 + }, + { + "epoch": 0.11702049690958395, + "grad_norm": 0.00896360632032156, + "learning_rate": 0.2973563164534908, + "loss": 0.1502, + "num_input_tokens_seen": 3106528, + "step": 2395 + }, + { + "epoch": 0.11726479857327829, + "grad_norm": 0.004499845206737518, + "learning_rate": 0.29734529468896287, + "loss": 0.1409, + "num_input_tokens_seen": 3112928, + "step": 2400 + }, + { + "epoch": 0.11726479857327829, + "eval_loss": 0.14451265335083008, + "eval_runtime": 402.3243, + "eval_samples_per_second": 90.437, + "eval_steps_per_second": 22.611, + "num_input_tokens_seen": 3112928, + "step": 2400 + }, + { + "epoch": 0.11750910023697261, + "grad_norm": 0.007701903115957975, + "learning_rate": 0.2973342502019388, + "loss": 0.1457, + "num_input_tokens_seen": 3119904, + "step": 2405 + }, + { + "epoch": 0.11775340190066694, + "grad_norm": 0.012430422939360142, + "learning_rate": 0.2973231829941219, + "loss": 0.1355, + "num_input_tokens_seen": 3126336, + "step": 2410 + }, + { + "epoch": 0.11799770356436128, + "grad_norm": 0.006526260636746883, + "learning_rate": 0.2973120930672188, + "loss": 0.14, + "num_input_tokens_seen": 3132352, + "step": 2415 + }, + { + "epoch": 0.1182420052280556, + "grad_norm": 0.013535221107304096, + "learning_rate": 0.2973009804229397, + "loss": 0.1449, + "num_input_tokens_seen": 3138560, + "step": 2420 + }, + { + "epoch": 0.11848630689174994, + "grad_norm": 0.008780235424637794, + "learning_rate": 0.29728984506299827, + "loss": 0.1618, + "num_input_tokens_seen": 3145120, + "step": 2425 + }, + { + "epoch": 0.11873060855544426, + "grad_norm": 0.01464722491800785, + "learning_rate": 0.2972786869891118, + "loss": 0.1907, + "num_input_tokens_seen": 3151776, + "step": 2430 + }, + { + "epoch": 0.11897491021913859, + "grad_norm": 0.006312259938567877, + "learning_rate": 0.29726750620300096, + "loss": 0.1585, + "num_input_tokens_seen": 3158496, + "step": 2435 + }, + { + "epoch": 0.11921921188283292, + "grad_norm": 0.0041227880865335464, + "learning_rate": 0.29725630270639003, + "loss": 0.1365, + "num_input_tokens_seen": 3164640, + "step": 2440 + }, + { + "epoch": 0.11946351354652725, + "grad_norm": 0.005052085965871811, + "learning_rate": 0.2972450765010067, + "loss": 0.1443, + "num_input_tokens_seen": 3170784, + "step": 2445 + }, + { + "epoch": 0.11970781521022159, + "grad_norm": 0.003872581524774432, + "learning_rate": 0.29723382758858213, + "loss": 0.1256, + "num_input_tokens_seen": 3177088, + "step": 2450 + }, + { + "epoch": 0.11995211687391591, + "grad_norm": 0.00572273088619113, + "learning_rate": 0.29722255597085107, + "loss": 0.1264, + "num_input_tokens_seen": 3183776, + "step": 2455 + }, + { + "epoch": 0.12019641853761025, + "grad_norm": 0.011160558089613914, + "learning_rate": 0.2972112616495518, + "loss": 0.1373, + "num_input_tokens_seen": 3190368, + "step": 2460 + }, + { + "epoch": 0.12044072020130457, + "grad_norm": 0.006367525551468134, + "learning_rate": 0.297199944626426, + "loss": 0.1244, + "num_input_tokens_seen": 3196384, + "step": 2465 + }, + { + "epoch": 0.1206850218649989, + "grad_norm": 0.012586004100739956, + "learning_rate": 0.2971886049032189, + "loss": 0.1519, + "num_input_tokens_seen": 3202944, + "step": 2470 + }, + { + "epoch": 0.12092932352869323, + "grad_norm": 0.011021323502063751, + "learning_rate": 0.29717724248167926, + "loss": 0.1704, + "num_input_tokens_seen": 3209376, + "step": 2475 + }, + { + "epoch": 0.12117362519238756, + "grad_norm": 0.0054773082956671715, + "learning_rate": 0.29716585736355927, + "loss": 0.143, + "num_input_tokens_seen": 3215744, + "step": 2480 + }, + { + "epoch": 0.1214179268560819, + "grad_norm": 0.005812338087707758, + "learning_rate": 0.2971544495506147, + "loss": 0.1414, + "num_input_tokens_seen": 3222144, + "step": 2485 + }, + { + "epoch": 0.12166222851977622, + "grad_norm": 0.007533779367804527, + "learning_rate": 0.2971430190446048, + "loss": 0.1517, + "num_input_tokens_seen": 3228672, + "step": 2490 + }, + { + "epoch": 0.12190653018347054, + "grad_norm": 0.00619752611964941, + "learning_rate": 0.2971315658472921, + "loss": 0.1482, + "num_input_tokens_seen": 3235264, + "step": 2495 + }, + { + "epoch": 0.12215083184716488, + "grad_norm": 0.009815577417612076, + "learning_rate": 0.2971200899604431, + "loss": 0.1558, + "num_input_tokens_seen": 3241760, + "step": 2500 + }, + { + "epoch": 0.1223951335108592, + "grad_norm": 0.007113547995686531, + "learning_rate": 0.29710859138582735, + "loss": 0.1587, + "num_input_tokens_seen": 3248512, + "step": 2505 + }, + { + "epoch": 0.12263943517455354, + "grad_norm": 0.009656806476414204, + "learning_rate": 0.29709707012521813, + "loss": 0.1523, + "num_input_tokens_seen": 3254720, + "step": 2510 + }, + { + "epoch": 0.12288373683824787, + "grad_norm": 0.004486394114792347, + "learning_rate": 0.29708552618039213, + "loss": 0.1476, + "num_input_tokens_seen": 3261408, + "step": 2515 + }, + { + "epoch": 0.12312803850194219, + "grad_norm": 0.009489977732300758, + "learning_rate": 0.2970739595531296, + "loss": 0.1646, + "num_input_tokens_seen": 3267808, + "step": 2520 + }, + { + "epoch": 0.12337234016563653, + "grad_norm": 0.0076123131439089775, + "learning_rate": 0.2970623702452143, + "loss": 0.1467, + "num_input_tokens_seen": 3274176, + "step": 2525 + }, + { + "epoch": 0.12361664182933085, + "grad_norm": 0.010514471679925919, + "learning_rate": 0.2970507582584334, + "loss": 0.138, + "num_input_tokens_seen": 3281440, + "step": 2530 + }, + { + "epoch": 0.12386094349302519, + "grad_norm": 0.006461451295763254, + "learning_rate": 0.2970391235945776, + "loss": 0.1681, + "num_input_tokens_seen": 3288352, + "step": 2535 + }, + { + "epoch": 0.12410524515671952, + "grad_norm": 0.009115886874496937, + "learning_rate": 0.2970274662554412, + "loss": 0.1503, + "num_input_tokens_seen": 3294848, + "step": 2540 + }, + { + "epoch": 0.12434954682041385, + "grad_norm": 0.006735718343406916, + "learning_rate": 0.2970157862428218, + "loss": 0.1431, + "num_input_tokens_seen": 3301536, + "step": 2545 + }, + { + "epoch": 0.12459384848410818, + "grad_norm": 0.0069580296985805035, + "learning_rate": 0.2970040835585206, + "loss": 0.1286, + "num_input_tokens_seen": 3308000, + "step": 2550 + }, + { + "epoch": 0.1248381501478025, + "grad_norm": 0.005741797853261232, + "learning_rate": 0.2969923582043424, + "loss": 0.1374, + "num_input_tokens_seen": 3314528, + "step": 2555 + }, + { + "epoch": 0.12508245181149683, + "grad_norm": 0.006566356867551804, + "learning_rate": 0.2969806101820953, + "loss": 0.1394, + "num_input_tokens_seen": 3321248, + "step": 2560 + }, + { + "epoch": 0.12532675347519118, + "grad_norm": 0.006448323838412762, + "learning_rate": 0.2969688394935911, + "loss": 0.1372, + "num_input_tokens_seen": 3327680, + "step": 2565 + }, + { + "epoch": 0.1255710551388855, + "grad_norm": 0.006665313150733709, + "learning_rate": 0.2969570461406449, + "loss": 0.1549, + "num_input_tokens_seen": 3334016, + "step": 2570 + }, + { + "epoch": 0.12581535680257983, + "grad_norm": 0.0048906030133366585, + "learning_rate": 0.29694523012507534, + "loss": 0.1403, + "num_input_tokens_seen": 3340960, + "step": 2575 + }, + { + "epoch": 0.12605965846627415, + "grad_norm": 0.005697744898498058, + "learning_rate": 0.2969333914487048, + "loss": 0.1619, + "num_input_tokens_seen": 3347488, + "step": 2580 + }, + { + "epoch": 0.12630396012996847, + "grad_norm": 0.01195291057229042, + "learning_rate": 0.2969215301133587, + "loss": 0.1477, + "num_input_tokens_seen": 3353792, + "step": 2585 + }, + { + "epoch": 0.12654826179366283, + "grad_norm": 0.006122096441686153, + "learning_rate": 0.29690964612086634, + "loss": 0.1169, + "num_input_tokens_seen": 3360864, + "step": 2590 + }, + { + "epoch": 0.12679256345735715, + "grad_norm": 0.006548020057380199, + "learning_rate": 0.2968977394730604, + "loss": 0.1513, + "num_input_tokens_seen": 3367680, + "step": 2595 + }, + { + "epoch": 0.12703686512105147, + "grad_norm": 0.00534789077937603, + "learning_rate": 0.296885810171777, + "loss": 0.1626, + "num_input_tokens_seen": 3374048, + "step": 2600 + }, + { + "epoch": 0.12703686512105147, + "eval_loss": 0.1462915986776352, + "eval_runtime": 401.8253, + "eval_samples_per_second": 90.549, + "eval_steps_per_second": 22.639, + "num_input_tokens_seen": 3374048, + "step": 2600 + }, + { + "epoch": 0.1272811667847458, + "grad_norm": 0.006092721596360207, + "learning_rate": 0.2968738582188558, + "loss": 0.1411, + "num_input_tokens_seen": 3380288, + "step": 2605 + }, + { + "epoch": 0.12752546844844012, + "grad_norm": 0.0065406435169279575, + "learning_rate": 0.2968618836161399, + "loss": 0.1575, + "num_input_tokens_seen": 3387040, + "step": 2610 + }, + { + "epoch": 0.12776977011213447, + "grad_norm": 0.007258173543959856, + "learning_rate": 0.296849886365476, + "loss": 0.1152, + "num_input_tokens_seen": 3393920, + "step": 2615 + }, + { + "epoch": 0.1280140717758288, + "grad_norm": 0.004331282339990139, + "learning_rate": 0.2968378664687142, + "loss": 0.1399, + "num_input_tokens_seen": 3400896, + "step": 2620 + }, + { + "epoch": 0.12825837343952312, + "grad_norm": 0.0036824143026024103, + "learning_rate": 0.296825823927708, + "loss": 0.1205, + "num_input_tokens_seen": 3407392, + "step": 2625 + }, + { + "epoch": 0.12850267510321745, + "grad_norm": 0.011034672148525715, + "learning_rate": 0.29681375874431476, + "loss": 0.1226, + "num_input_tokens_seen": 3414496, + "step": 2630 + }, + { + "epoch": 0.12874697676691177, + "grad_norm": 0.008566209115087986, + "learning_rate": 0.29680167092039483, + "loss": 0.1435, + "num_input_tokens_seen": 3420800, + "step": 2635 + }, + { + "epoch": 0.12899127843060612, + "grad_norm": 0.00572593929246068, + "learning_rate": 0.2967895604578125, + "loss": 0.1495, + "num_input_tokens_seen": 3427712, + "step": 2640 + }, + { + "epoch": 0.12923558009430045, + "grad_norm": 0.0051056863740086555, + "learning_rate": 0.2967774273584352, + "loss": 0.141, + "num_input_tokens_seen": 3434400, + "step": 2645 + }, + { + "epoch": 0.12947988175799477, + "grad_norm": 0.007015488110482693, + "learning_rate": 0.2967652716241342, + "loss": 0.1269, + "num_input_tokens_seen": 3441088, + "step": 2650 + }, + { + "epoch": 0.1297241834216891, + "grad_norm": 0.005024552345275879, + "learning_rate": 0.29675309325678384, + "loss": 0.1734, + "num_input_tokens_seen": 3447968, + "step": 2655 + }, + { + "epoch": 0.12996848508538345, + "grad_norm": 0.004386928863823414, + "learning_rate": 0.29674089225826233, + "loss": 0.1348, + "num_input_tokens_seen": 3454080, + "step": 2660 + }, + { + "epoch": 0.13021278674907777, + "grad_norm": 0.004079080652445555, + "learning_rate": 0.29672866863045116, + "loss": 0.136, + "num_input_tokens_seen": 3460608, + "step": 2665 + }, + { + "epoch": 0.1304570884127721, + "grad_norm": 0.006047117058187723, + "learning_rate": 0.2967164223752354, + "loss": 0.1486, + "num_input_tokens_seen": 3466976, + "step": 2670 + }, + { + "epoch": 0.13070139007646642, + "grad_norm": 0.007762869819998741, + "learning_rate": 0.2967041534945035, + "loss": 0.1518, + "num_input_tokens_seen": 3473632, + "step": 2675 + }, + { + "epoch": 0.13094569174016074, + "grad_norm": 0.006465918384492397, + "learning_rate": 0.2966918619901476, + "loss": 0.1493, + "num_input_tokens_seen": 3480160, + "step": 2680 + }, + { + "epoch": 0.1311899934038551, + "grad_norm": 0.0057994830422103405, + "learning_rate": 0.2966795478640631, + "loss": 0.1389, + "num_input_tokens_seen": 3486784, + "step": 2685 + }, + { + "epoch": 0.13143429506754942, + "grad_norm": 0.0066549708135426044, + "learning_rate": 0.29666721111814903, + "loss": 0.1213, + "num_input_tokens_seen": 3493376, + "step": 2690 + }, + { + "epoch": 0.13167859673124374, + "grad_norm": 0.0063249277882277966, + "learning_rate": 0.2966548517543079, + "loss": 0.1118, + "num_input_tokens_seen": 3500128, + "step": 2695 + }, + { + "epoch": 0.13192289839493807, + "grad_norm": 0.006085045635700226, + "learning_rate": 0.29664246977444564, + "loss": 0.1305, + "num_input_tokens_seen": 3507264, + "step": 2700 + }, + { + "epoch": 0.1321672000586324, + "grad_norm": 0.0069174147211015224, + "learning_rate": 0.2966300651804717, + "loss": 0.1394, + "num_input_tokens_seen": 3513664, + "step": 2705 + }, + { + "epoch": 0.13241150172232674, + "grad_norm": 0.008866139687597752, + "learning_rate": 0.296617637974299, + "loss": 0.1478, + "num_input_tokens_seen": 3520256, + "step": 2710 + }, + { + "epoch": 0.13265580338602107, + "grad_norm": 0.007250121794641018, + "learning_rate": 0.2966051881578441, + "loss": 0.134, + "num_input_tokens_seen": 3526272, + "step": 2715 + }, + { + "epoch": 0.1329001050497154, + "grad_norm": 0.006380570121109486, + "learning_rate": 0.29659271573302676, + "loss": 0.1308, + "num_input_tokens_seen": 3532672, + "step": 2720 + }, + { + "epoch": 0.1331444067134097, + "grad_norm": 0.008857816457748413, + "learning_rate": 0.2965802207017705, + "loss": 0.1622, + "num_input_tokens_seen": 3538688, + "step": 2725 + }, + { + "epoch": 0.13338870837710404, + "grad_norm": 0.007466928102076054, + "learning_rate": 0.2965677030660021, + "loss": 0.1339, + "num_input_tokens_seen": 3545056, + "step": 2730 + }, + { + "epoch": 0.1336330100407984, + "grad_norm": 0.006451708264648914, + "learning_rate": 0.2965551628276521, + "loss": 0.1474, + "num_input_tokens_seen": 3551584, + "step": 2735 + }, + { + "epoch": 0.1338773117044927, + "grad_norm": 0.00776198785752058, + "learning_rate": 0.29654259998865423, + "loss": 0.1596, + "num_input_tokens_seen": 3557888, + "step": 2740 + }, + { + "epoch": 0.13412161336818704, + "grad_norm": 0.006708407308906317, + "learning_rate": 0.2965300145509458, + "loss": 0.1294, + "num_input_tokens_seen": 3564704, + "step": 2745 + }, + { + "epoch": 0.13436591503188136, + "grad_norm": 0.01115484070032835, + "learning_rate": 0.2965174065164678, + "loss": 0.1627, + "num_input_tokens_seen": 3571744, + "step": 2750 + }, + { + "epoch": 0.13461021669557569, + "grad_norm": 0.010141697712242603, + "learning_rate": 0.2965047758871644, + "loss": 0.15, + "num_input_tokens_seen": 3578048, + "step": 2755 + }, + { + "epoch": 0.13485451835927004, + "grad_norm": 0.0058759599924087524, + "learning_rate": 0.2964921226649835, + "loss": 0.1705, + "num_input_tokens_seen": 3584416, + "step": 2760 + }, + { + "epoch": 0.13509882002296436, + "grad_norm": 0.007786884438246489, + "learning_rate": 0.2964794468518763, + "loss": 0.149, + "num_input_tokens_seen": 3591168, + "step": 2765 + }, + { + "epoch": 0.13534312168665869, + "grad_norm": 0.005831062328070402, + "learning_rate": 0.2964667484497977, + "loss": 0.1551, + "num_input_tokens_seen": 3598240, + "step": 2770 + }, + { + "epoch": 0.135587423350353, + "grad_norm": 0.005583528894931078, + "learning_rate": 0.29645402746070587, + "loss": 0.1487, + "num_input_tokens_seen": 3604512, + "step": 2775 + }, + { + "epoch": 0.13583172501404733, + "grad_norm": 0.00941088330000639, + "learning_rate": 0.2964412838865625, + "loss": 0.1648, + "num_input_tokens_seen": 3610688, + "step": 2780 + }, + { + "epoch": 0.13607602667774168, + "grad_norm": 0.0063462285324931145, + "learning_rate": 0.29642851772933293, + "loss": 0.1454, + "num_input_tokens_seen": 3617056, + "step": 2785 + }, + { + "epoch": 0.136320328341436, + "grad_norm": 0.003491115989163518, + "learning_rate": 0.29641572899098567, + "loss": 0.1354, + "num_input_tokens_seen": 3623872, + "step": 2790 + }, + { + "epoch": 0.13656463000513033, + "grad_norm": 0.005101524293422699, + "learning_rate": 0.29640291767349314, + "loss": 0.1219, + "num_input_tokens_seen": 3630976, + "step": 2795 + }, + { + "epoch": 0.13680893166882466, + "grad_norm": 0.0058134617283940315, + "learning_rate": 0.2963900837788308, + "loss": 0.117, + "num_input_tokens_seen": 3637152, + "step": 2800 + }, + { + "epoch": 0.13680893166882466, + "eval_loss": 0.14901645481586456, + "eval_runtime": 401.8883, + "eval_samples_per_second": 90.535, + "eval_steps_per_second": 22.636, + "num_input_tokens_seen": 3637152, + "step": 2800 + }, + { + "epoch": 0.13705323333251898, + "grad_norm": 0.005075949244201183, + "learning_rate": 0.2963772273089779, + "loss": 0.1569, + "num_input_tokens_seen": 3643488, + "step": 2805 + }, + { + "epoch": 0.13729753499621333, + "grad_norm": 0.005599355325102806, + "learning_rate": 0.2963643482659171, + "loss": 0.1293, + "num_input_tokens_seen": 3650144, + "step": 2810 + }, + { + "epoch": 0.13754183665990766, + "grad_norm": 0.007651086430996656, + "learning_rate": 0.2963514466516345, + "loss": 0.1747, + "num_input_tokens_seen": 3656544, + "step": 2815 + }, + { + "epoch": 0.13778613832360198, + "grad_norm": 0.007631708402186632, + "learning_rate": 0.2963385224681196, + "loss": 0.162, + "num_input_tokens_seen": 3662848, + "step": 2820 + }, + { + "epoch": 0.1380304399872963, + "grad_norm": 0.0038130036555230618, + "learning_rate": 0.29632557571736556, + "loss": 0.1321, + "num_input_tokens_seen": 3669696, + "step": 2825 + }, + { + "epoch": 0.13827474165099066, + "grad_norm": 0.0057632168754935265, + "learning_rate": 0.2963126064013689, + "loss": 0.1638, + "num_input_tokens_seen": 3676096, + "step": 2830 + }, + { + "epoch": 0.13851904331468498, + "grad_norm": 0.005079422611743212, + "learning_rate": 0.29629961452212966, + "loss": 0.1175, + "num_input_tokens_seen": 3682656, + "step": 2835 + }, + { + "epoch": 0.1387633449783793, + "grad_norm": 0.010958133265376091, + "learning_rate": 0.2962866000816513, + "loss": 0.1429, + "num_input_tokens_seen": 3688928, + "step": 2840 + }, + { + "epoch": 0.13900764664207363, + "grad_norm": 0.006899616681039333, + "learning_rate": 0.2962735630819409, + "loss": 0.1612, + "num_input_tokens_seen": 3695840, + "step": 2845 + }, + { + "epoch": 0.13925194830576795, + "grad_norm": 0.004329555202275515, + "learning_rate": 0.2962605035250089, + "loss": 0.13, + "num_input_tokens_seen": 3702080, + "step": 2850 + }, + { + "epoch": 0.1394962499694623, + "grad_norm": 0.00706408079713583, + "learning_rate": 0.29624742141286914, + "loss": 0.1606, + "num_input_tokens_seen": 3708928, + "step": 2855 + }, + { + "epoch": 0.13974055163315663, + "grad_norm": 0.0069637601263821125, + "learning_rate": 0.29623431674753925, + "loss": 0.1067, + "num_input_tokens_seen": 3715648, + "step": 2860 + }, + { + "epoch": 0.13998485329685095, + "grad_norm": 0.012289231643080711, + "learning_rate": 0.29622118953103993, + "loss": 0.1458, + "num_input_tokens_seen": 3721984, + "step": 2865 + }, + { + "epoch": 0.14022915496054528, + "grad_norm": 0.007180750835686922, + "learning_rate": 0.2962080397653957, + "loss": 0.1647, + "num_input_tokens_seen": 3728384, + "step": 2870 + }, + { + "epoch": 0.1404734566242396, + "grad_norm": 0.007220921106636524, + "learning_rate": 0.29619486745263435, + "loss": 0.1607, + "num_input_tokens_seen": 3734656, + "step": 2875 + }, + { + "epoch": 0.14071775828793395, + "grad_norm": 0.004536958411335945, + "learning_rate": 0.2961816725947873, + "loss": 0.127, + "num_input_tokens_seen": 3741120, + "step": 2880 + }, + { + "epoch": 0.14096205995162828, + "grad_norm": 0.010077732615172863, + "learning_rate": 0.29616845519388924, + "loss": 0.1315, + "num_input_tokens_seen": 3747488, + "step": 2885 + }, + { + "epoch": 0.1412063616153226, + "grad_norm": 0.008811172097921371, + "learning_rate": 0.2961552152519785, + "loss": 0.1257, + "num_input_tokens_seen": 3754176, + "step": 2890 + }, + { + "epoch": 0.14145066327901692, + "grad_norm": 0.006412859074771404, + "learning_rate": 0.29614195277109695, + "loss": 0.1477, + "num_input_tokens_seen": 3760128, + "step": 2895 + }, + { + "epoch": 0.14169496494271125, + "grad_norm": 0.006547005847096443, + "learning_rate": 0.2961286677532897, + "loss": 0.128, + "num_input_tokens_seen": 3766688, + "step": 2900 + }, + { + "epoch": 0.1419392666064056, + "grad_norm": 0.00698671443387866, + "learning_rate": 0.2961153602006055, + "loss": 0.1216, + "num_input_tokens_seen": 3773024, + "step": 2905 + }, + { + "epoch": 0.14218356827009992, + "grad_norm": 0.012459772638976574, + "learning_rate": 0.29610203011509656, + "loss": 0.1606, + "num_input_tokens_seen": 3779328, + "step": 2910 + }, + { + "epoch": 0.14242786993379425, + "grad_norm": 0.009354677051305771, + "learning_rate": 0.29608867749881856, + "loss": 0.1573, + "num_input_tokens_seen": 3785248, + "step": 2915 + }, + { + "epoch": 0.14267217159748857, + "grad_norm": 0.003984795417636633, + "learning_rate": 0.29607530235383067, + "loss": 0.1134, + "num_input_tokens_seen": 3791552, + "step": 2920 + }, + { + "epoch": 0.1429164732611829, + "grad_norm": 0.007021724712103605, + "learning_rate": 0.2960619046821954, + "loss": 0.1408, + "num_input_tokens_seen": 3797952, + "step": 2925 + }, + { + "epoch": 0.14316077492487725, + "grad_norm": 0.006713752169162035, + "learning_rate": 0.2960484844859789, + "loss": 0.1393, + "num_input_tokens_seen": 3804064, + "step": 2930 + }, + { + "epoch": 0.14340507658857157, + "grad_norm": 0.007559201214462519, + "learning_rate": 0.29603504176725076, + "loss": 0.1493, + "num_input_tokens_seen": 3810880, + "step": 2935 + }, + { + "epoch": 0.1436493782522659, + "grad_norm": 0.0070272330194711685, + "learning_rate": 0.296021576528084, + "loss": 0.1396, + "num_input_tokens_seen": 3817600, + "step": 2940 + }, + { + "epoch": 0.14389367991596022, + "grad_norm": 0.004861912690103054, + "learning_rate": 0.29600808877055507, + "loss": 0.1539, + "num_input_tokens_seen": 3824160, + "step": 2945 + }, + { + "epoch": 0.14413798157965454, + "grad_norm": 0.005469650495797396, + "learning_rate": 0.29599457849674404, + "loss": 0.1384, + "num_input_tokens_seen": 3830528, + "step": 2950 + }, + { + "epoch": 0.1443822832433489, + "grad_norm": 0.006416147109121084, + "learning_rate": 0.2959810457087343, + "loss": 0.1217, + "num_input_tokens_seen": 3837120, + "step": 2955 + }, + { + "epoch": 0.14462658490704322, + "grad_norm": 0.01056357380002737, + "learning_rate": 0.2959674904086128, + "loss": 0.1525, + "num_input_tokens_seen": 3843968, + "step": 2960 + }, + { + "epoch": 0.14487088657073754, + "grad_norm": 0.004586751572787762, + "learning_rate": 0.2959539125984699, + "loss": 0.1218, + "num_input_tokens_seen": 3850400, + "step": 2965 + }, + { + "epoch": 0.14511518823443187, + "grad_norm": 0.017131291329860687, + "learning_rate": 0.2959403122803996, + "loss": 0.1589, + "num_input_tokens_seen": 3856896, + "step": 2970 + }, + { + "epoch": 0.1453594898981262, + "grad_norm": 0.011254126206040382, + "learning_rate": 0.2959266894564991, + "loss": 0.1432, + "num_input_tokens_seen": 3863520, + "step": 2975 + }, + { + "epoch": 0.14560379156182054, + "grad_norm": 0.006342659704387188, + "learning_rate": 0.2959130441288692, + "loss": 0.1196, + "num_input_tokens_seen": 3870272, + "step": 2980 + }, + { + "epoch": 0.14584809322551487, + "grad_norm": 0.005669768899679184, + "learning_rate": 0.2958993762996143, + "loss": 0.1224, + "num_input_tokens_seen": 3877088, + "step": 2985 + }, + { + "epoch": 0.1460923948892092, + "grad_norm": 0.008901184424757957, + "learning_rate": 0.2958856859708421, + "loss": 0.1502, + "num_input_tokens_seen": 3883776, + "step": 2990 + }, + { + "epoch": 0.14633669655290352, + "grad_norm": 0.005730160512030125, + "learning_rate": 0.2958719731446638, + "loss": 0.1672, + "num_input_tokens_seen": 3890112, + "step": 2995 + }, + { + "epoch": 0.14658099821659787, + "grad_norm": 0.004826069809496403, + "learning_rate": 0.29585823782319404, + "loss": 0.1551, + "num_input_tokens_seen": 3896512, + "step": 3000 + }, + { + "epoch": 0.14658099821659787, + "eval_loss": 0.15010009706020355, + "eval_runtime": 402.2646, + "eval_samples_per_second": 90.45, + "eval_steps_per_second": 22.614, + "num_input_tokens_seen": 3896512, + "step": 3000 + }, + { + "epoch": 0.1468252998802922, + "grad_norm": 0.0042345342226326466, + "learning_rate": 0.2958444800085511, + "loss": 0.1612, + "num_input_tokens_seen": 3902688, + "step": 3005 + }, + { + "epoch": 0.14706960154398652, + "grad_norm": 0.0038030026480555534, + "learning_rate": 0.2958306997028565, + "loss": 0.134, + "num_input_tokens_seen": 3909408, + "step": 3010 + }, + { + "epoch": 0.14731390320768084, + "grad_norm": 0.003873246256262064, + "learning_rate": 0.2958168969082354, + "loss": 0.153, + "num_input_tokens_seen": 3915488, + "step": 3015 + }, + { + "epoch": 0.14755820487137516, + "grad_norm": 0.005529769696295261, + "learning_rate": 0.2958030716268164, + "loss": 0.1515, + "num_input_tokens_seen": 3921632, + "step": 3020 + }, + { + "epoch": 0.14780250653506952, + "grad_norm": 0.006268240045756102, + "learning_rate": 0.2957892238607314, + "loss": 0.1466, + "num_input_tokens_seen": 3928640, + "step": 3025 + }, + { + "epoch": 0.14804680819876384, + "grad_norm": 0.007846810854971409, + "learning_rate": 0.2957753536121161, + "loss": 0.1452, + "num_input_tokens_seen": 3935200, + "step": 3030 + }, + { + "epoch": 0.14829110986245816, + "grad_norm": 0.005251065827906132, + "learning_rate": 0.29576146088310923, + "loss": 0.1435, + "num_input_tokens_seen": 3941856, + "step": 3035 + }, + { + "epoch": 0.1485354115261525, + "grad_norm": 0.004493350628763437, + "learning_rate": 0.2957475456758533, + "loss": 0.1256, + "num_input_tokens_seen": 3948640, + "step": 3040 + }, + { + "epoch": 0.1487797131898468, + "grad_norm": 0.0070021250285208225, + "learning_rate": 0.2957336079924944, + "loss": 0.1424, + "num_input_tokens_seen": 3955264, + "step": 3045 + }, + { + "epoch": 0.14902401485354116, + "grad_norm": 0.012716337107121944, + "learning_rate": 0.2957196478351816, + "loss": 0.1583, + "num_input_tokens_seen": 3961504, + "step": 3050 + }, + { + "epoch": 0.1492683165172355, + "grad_norm": 0.00458125676959753, + "learning_rate": 0.295705665206068, + "loss": 0.1343, + "num_input_tokens_seen": 3967424, + "step": 3055 + }, + { + "epoch": 0.1495126181809298, + "grad_norm": 0.00432152533903718, + "learning_rate": 0.2956916601073097, + "loss": 0.1042, + "num_input_tokens_seen": 3974336, + "step": 3060 + }, + { + "epoch": 0.14975691984462414, + "grad_norm": 0.007240593899041414, + "learning_rate": 0.29567763254106655, + "loss": 0.1552, + "num_input_tokens_seen": 3980832, + "step": 3065 + }, + { + "epoch": 0.15000122150831846, + "grad_norm": 0.02533837966620922, + "learning_rate": 0.29566358250950175, + "loss": 0.186, + "num_input_tokens_seen": 3987008, + "step": 3070 + }, + { + "epoch": 0.1502455231720128, + "grad_norm": 0.011921662837266922, + "learning_rate": 0.295649510014782, + "loss": 0.1628, + "num_input_tokens_seen": 3993728, + "step": 3075 + }, + { + "epoch": 0.15048982483570714, + "grad_norm": 0.007870647124946117, + "learning_rate": 0.2956354150590775, + "loss": 0.1462, + "num_input_tokens_seen": 4000288, + "step": 3080 + }, + { + "epoch": 0.15073412649940146, + "grad_norm": 0.007252035662531853, + "learning_rate": 0.2956212976445618, + "loss": 0.1267, + "num_input_tokens_seen": 4006944, + "step": 3085 + }, + { + "epoch": 0.15097842816309578, + "grad_norm": 0.005040792282670736, + "learning_rate": 0.295607157773412, + "loss": 0.1384, + "num_input_tokens_seen": 4013472, + "step": 3090 + }, + { + "epoch": 0.1512227298267901, + "grad_norm": 0.005427815020084381, + "learning_rate": 0.2955929954478087, + "loss": 0.1691, + "num_input_tokens_seen": 4019712, + "step": 3095 + }, + { + "epoch": 0.15146703149048446, + "grad_norm": 0.008965421468019485, + "learning_rate": 0.29557881066993585, + "loss": 0.1212, + "num_input_tokens_seen": 4026240, + "step": 3100 + }, + { + "epoch": 0.15171133315417878, + "grad_norm": 0.005682289134711027, + "learning_rate": 0.29556460344198093, + "loss": 0.133, + "num_input_tokens_seen": 4032800, + "step": 3105 + }, + { + "epoch": 0.1519556348178731, + "grad_norm": 0.012129625305533409, + "learning_rate": 0.29555037376613486, + "loss": 0.1538, + "num_input_tokens_seen": 4039712, + "step": 3110 + }, + { + "epoch": 0.15219993648156743, + "grad_norm": 0.019299698993563652, + "learning_rate": 0.29553612164459203, + "loss": 0.1686, + "num_input_tokens_seen": 4045792, + "step": 3115 + }, + { + "epoch": 0.15244423814526176, + "grad_norm": 0.007198733743280172, + "learning_rate": 0.29552184707955037, + "loss": 0.1569, + "num_input_tokens_seen": 4052384, + "step": 3120 + }, + { + "epoch": 0.1526885398089561, + "grad_norm": 0.006550279445946217, + "learning_rate": 0.29550755007321117, + "loss": 0.1595, + "num_input_tokens_seen": 4058720, + "step": 3125 + }, + { + "epoch": 0.15293284147265043, + "grad_norm": 0.004274212289601564, + "learning_rate": 0.29549323062777916, + "loss": 0.1416, + "num_input_tokens_seen": 4065344, + "step": 3130 + }, + { + "epoch": 0.15317714313634476, + "grad_norm": 0.003892359556630254, + "learning_rate": 0.29547888874546263, + "loss": 0.1445, + "num_input_tokens_seen": 4071712, + "step": 3135 + }, + { + "epoch": 0.15342144480003908, + "grad_norm": 0.011292497627437115, + "learning_rate": 0.2954645244284732, + "loss": 0.1466, + "num_input_tokens_seen": 4078528, + "step": 3140 + }, + { + "epoch": 0.1536657464637334, + "grad_norm": 0.007157387211918831, + "learning_rate": 0.2954501376790261, + "loss": 0.1357, + "num_input_tokens_seen": 4085056, + "step": 3145 + }, + { + "epoch": 0.15391004812742776, + "grad_norm": 0.004868432879447937, + "learning_rate": 0.29543572849933997, + "loss": 0.1569, + "num_input_tokens_seen": 4091296, + "step": 3150 + }, + { + "epoch": 0.15415434979112208, + "grad_norm": 0.007058338262140751, + "learning_rate": 0.2954212968916368, + "loss": 0.1448, + "num_input_tokens_seen": 4097696, + "step": 3155 + }, + { + "epoch": 0.1543986514548164, + "grad_norm": 0.003827763954177499, + "learning_rate": 0.29540684285814217, + "loss": 0.1474, + "num_input_tokens_seen": 4104000, + "step": 3160 + }, + { + "epoch": 0.15464295311851073, + "grad_norm": 0.007140215951949358, + "learning_rate": 0.2953923664010851, + "loss": 0.1398, + "num_input_tokens_seen": 4110272, + "step": 3165 + }, + { + "epoch": 0.15488725478220508, + "grad_norm": 0.004550101701170206, + "learning_rate": 0.295377867522698, + "loss": 0.1353, + "num_input_tokens_seen": 4117024, + "step": 3170 + }, + { + "epoch": 0.1551315564458994, + "grad_norm": 0.004532177001237869, + "learning_rate": 0.2953633462252168, + "loss": 0.1964, + "num_input_tokens_seen": 4123360, + "step": 3175 + }, + { + "epoch": 0.15537585810959373, + "grad_norm": 0.01317333523184061, + "learning_rate": 0.2953488025108809, + "loss": 0.1715, + "num_input_tokens_seen": 4129760, + "step": 3180 + }, + { + "epoch": 0.15562015977328805, + "grad_norm": 0.004455049522221088, + "learning_rate": 0.295334236381933, + "loss": 0.1309, + "num_input_tokens_seen": 4136832, + "step": 3185 + }, + { + "epoch": 0.15586446143698238, + "grad_norm": 0.006748960819095373, + "learning_rate": 0.29531964784061954, + "loss": 0.2001, + "num_input_tokens_seen": 4143264, + "step": 3190 + }, + { + "epoch": 0.15610876310067673, + "grad_norm": 0.009004302322864532, + "learning_rate": 0.2953050368891902, + "loss": 0.1501, + "num_input_tokens_seen": 4150656, + "step": 3195 + }, + { + "epoch": 0.15635306476437105, + "grad_norm": 0.005905374884605408, + "learning_rate": 0.29529040352989805, + "loss": 0.1438, + "num_input_tokens_seen": 4157216, + "step": 3200 + }, + { + "epoch": 0.15635306476437105, + "eval_loss": 0.146976038813591, + "eval_runtime": 402.2321, + "eval_samples_per_second": 90.458, + "eval_steps_per_second": 22.616, + "num_input_tokens_seen": 4157216, + "step": 3200 + }, + { + "epoch": 0.15659736642806538, + "grad_norm": 0.004526370670646429, + "learning_rate": 0.29527574776499993, + "loss": 0.1122, + "num_input_tokens_seen": 4163808, + "step": 3205 + }, + { + "epoch": 0.1568416680917597, + "grad_norm": 0.008717632852494717, + "learning_rate": 0.2952610695967558, + "loss": 0.1772, + "num_input_tokens_seen": 4170240, + "step": 3210 + }, + { + "epoch": 0.15708596975545402, + "grad_norm": 0.00574049586430192, + "learning_rate": 0.29524636902742935, + "loss": 0.1437, + "num_input_tokens_seen": 4177376, + "step": 3215 + }, + { + "epoch": 0.15733027141914838, + "grad_norm": 0.007429580669850111, + "learning_rate": 0.2952316460592875, + "loss": 0.1429, + "num_input_tokens_seen": 4184032, + "step": 3220 + }, + { + "epoch": 0.1575745730828427, + "grad_norm": 0.012980662286281586, + "learning_rate": 0.29521690069460066, + "loss": 0.1671, + "num_input_tokens_seen": 4190304, + "step": 3225 + }, + { + "epoch": 0.15781887474653702, + "grad_norm": 0.00748571427538991, + "learning_rate": 0.29520213293564285, + "loss": 0.1405, + "num_input_tokens_seen": 4196960, + "step": 3230 + }, + { + "epoch": 0.15806317641023135, + "grad_norm": 0.00416966388002038, + "learning_rate": 0.29518734278469144, + "loss": 0.1347, + "num_input_tokens_seen": 4203552, + "step": 3235 + }, + { + "epoch": 0.15830747807392567, + "grad_norm": 0.005565917119383812, + "learning_rate": 0.29517253024402723, + "loss": 0.152, + "num_input_tokens_seen": 4209696, + "step": 3240 + }, + { + "epoch": 0.15855177973762002, + "grad_norm": 0.00484222499653697, + "learning_rate": 0.2951576953159345, + "loss": 0.1085, + "num_input_tokens_seen": 4216128, + "step": 3245 + }, + { + "epoch": 0.15879608140131435, + "grad_norm": 0.0063702804036438465, + "learning_rate": 0.29514283800270097, + "loss": 0.1433, + "num_input_tokens_seen": 4222912, + "step": 3250 + }, + { + "epoch": 0.15904038306500867, + "grad_norm": 0.011864092200994492, + "learning_rate": 0.2951279583066179, + "loss": 0.1349, + "num_input_tokens_seen": 4229792, + "step": 3255 + }, + { + "epoch": 0.159284684728703, + "grad_norm": 0.007720791269093752, + "learning_rate": 0.2951130562299798, + "loss": 0.1375, + "num_input_tokens_seen": 4236384, + "step": 3260 + }, + { + "epoch": 0.15952898639239732, + "grad_norm": 0.006543894298374653, + "learning_rate": 0.29509813177508487, + "loss": 0.157, + "num_input_tokens_seen": 4243488, + "step": 3265 + }, + { + "epoch": 0.15977328805609167, + "grad_norm": 0.009356270544230938, + "learning_rate": 0.2950831849442346, + "loss": 0.1365, + "num_input_tokens_seen": 4250112, + "step": 3270 + }, + { + "epoch": 0.160017589719786, + "grad_norm": 0.003859233809635043, + "learning_rate": 0.2950682157397339, + "loss": 0.122, + "num_input_tokens_seen": 4256512, + "step": 3275 + }, + { + "epoch": 0.16026189138348032, + "grad_norm": 0.007133918348699808, + "learning_rate": 0.2950532241638914, + "loss": 0.1249, + "num_input_tokens_seen": 4263264, + "step": 3280 + }, + { + "epoch": 0.16050619304717464, + "grad_norm": 0.007374327164143324, + "learning_rate": 0.2950382102190188, + "loss": 0.153, + "num_input_tokens_seen": 4269600, + "step": 3285 + }, + { + "epoch": 0.16075049471086897, + "grad_norm": 0.00995838176459074, + "learning_rate": 0.2950231739074316, + "loss": 0.1589, + "num_input_tokens_seen": 4275648, + "step": 3290 + }, + { + "epoch": 0.16099479637456332, + "grad_norm": 0.004716991912573576, + "learning_rate": 0.29500811523144843, + "loss": 0.1453, + "num_input_tokens_seen": 4281856, + "step": 3295 + }, + { + "epoch": 0.16123909803825764, + "grad_norm": 0.008197176270186901, + "learning_rate": 0.2949930341933917, + "loss": 0.1605, + "num_input_tokens_seen": 4287968, + "step": 3300 + }, + { + "epoch": 0.16148339970195197, + "grad_norm": 0.006398800760507584, + "learning_rate": 0.29497793079558693, + "loss": 0.1606, + "num_input_tokens_seen": 4295104, + "step": 3305 + }, + { + "epoch": 0.1617277013656463, + "grad_norm": 0.0040215603075921535, + "learning_rate": 0.2949628050403633, + "loss": 0.1314, + "num_input_tokens_seen": 4301440, + "step": 3310 + }, + { + "epoch": 0.16197200302934062, + "grad_norm": 0.005545100197196007, + "learning_rate": 0.2949476569300535, + "loss": 0.143, + "num_input_tokens_seen": 4307648, + "step": 3315 + }, + { + "epoch": 0.16221630469303497, + "grad_norm": 0.00582894217222929, + "learning_rate": 0.29493248646699344, + "loss": 0.1354, + "num_input_tokens_seen": 4313984, + "step": 3320 + }, + { + "epoch": 0.1624606063567293, + "grad_norm": 0.004599748179316521, + "learning_rate": 0.29491729365352265, + "loss": 0.15, + "num_input_tokens_seen": 4320832, + "step": 3325 + }, + { + "epoch": 0.16270490802042362, + "grad_norm": 0.005137608852237463, + "learning_rate": 0.29490207849198397, + "loss": 0.1469, + "num_input_tokens_seen": 4327104, + "step": 3330 + }, + { + "epoch": 0.16294920968411794, + "grad_norm": 0.005961100570857525, + "learning_rate": 0.29488684098472384, + "loss": 0.1552, + "num_input_tokens_seen": 4333952, + "step": 3335 + }, + { + "epoch": 0.1631935113478123, + "grad_norm": 0.006460412405431271, + "learning_rate": 0.2948715811340921, + "loss": 0.1478, + "num_input_tokens_seen": 4340480, + "step": 3340 + }, + { + "epoch": 0.16343781301150662, + "grad_norm": 0.004630151204764843, + "learning_rate": 0.294856298942442, + "loss": 0.1327, + "num_input_tokens_seen": 4347040, + "step": 3345 + }, + { + "epoch": 0.16368211467520094, + "grad_norm": 0.006064065266400576, + "learning_rate": 0.2948409944121302, + "loss": 0.1814, + "num_input_tokens_seen": 4352864, + "step": 3350 + }, + { + "epoch": 0.16392641633889526, + "grad_norm": 0.010582013987004757, + "learning_rate": 0.29482566754551687, + "loss": 0.1648, + "num_input_tokens_seen": 4359328, + "step": 3355 + }, + { + "epoch": 0.1641707180025896, + "grad_norm": 0.0030569855589419603, + "learning_rate": 0.2948103183449656, + "loss": 0.1181, + "num_input_tokens_seen": 4365856, + "step": 3360 + }, + { + "epoch": 0.16441501966628394, + "grad_norm": 0.006651660427451134, + "learning_rate": 0.2947949468128435, + "loss": 0.1524, + "num_input_tokens_seen": 4372320, + "step": 3365 + }, + { + "epoch": 0.16465932132997826, + "grad_norm": 0.009209196083247662, + "learning_rate": 0.2947795529515209, + "loss": 0.154, + "num_input_tokens_seen": 4378752, + "step": 3370 + }, + { + "epoch": 0.1649036229936726, + "grad_norm": 0.007435793522745371, + "learning_rate": 0.29476413676337193, + "loss": 0.1441, + "num_input_tokens_seen": 4385856, + "step": 3375 + }, + { + "epoch": 0.1651479246573669, + "grad_norm": 0.011264672502875328, + "learning_rate": 0.2947486982507738, + "loss": 0.1498, + "num_input_tokens_seen": 4392416, + "step": 3380 + }, + { + "epoch": 0.16539222632106124, + "grad_norm": 0.006896717473864555, + "learning_rate": 0.29473323741610735, + "loss": 0.1426, + "num_input_tokens_seen": 4399072, + "step": 3385 + }, + { + "epoch": 0.1656365279847556, + "grad_norm": 0.012486022897064686, + "learning_rate": 0.2947177542617569, + "loss": 0.1493, + "num_input_tokens_seen": 4405472, + "step": 3390 + }, + { + "epoch": 0.1658808296484499, + "grad_norm": 0.012338816188275814, + "learning_rate": 0.2947022487901101, + "loss": 0.1791, + "num_input_tokens_seen": 4411968, + "step": 3395 + }, + { + "epoch": 0.16612513131214424, + "grad_norm": 0.009308363310992718, + "learning_rate": 0.2946867210035581, + "loss": 0.1477, + "num_input_tokens_seen": 4418592, + "step": 3400 + }, + { + "epoch": 0.16612513131214424, + "eval_loss": 0.14681977033615112, + "eval_runtime": 402.1083, + "eval_samples_per_second": 90.486, + "eval_steps_per_second": 22.623, + "num_input_tokens_seen": 4418592, + "step": 3400 + }, + { + "epoch": 0.16636943297583856, + "grad_norm": 0.01494782604277134, + "learning_rate": 0.2946711709044954, + "loss": 0.1652, + "num_input_tokens_seen": 4425088, + "step": 3405 + }, + { + "epoch": 0.16661373463953288, + "grad_norm": 0.006853220518678427, + "learning_rate": 0.2946555984953202, + "loss": 0.141, + "num_input_tokens_seen": 4431584, + "step": 3410 + }, + { + "epoch": 0.16685803630322724, + "grad_norm": 0.005910870619118214, + "learning_rate": 0.2946400037784338, + "loss": 0.1697, + "num_input_tokens_seen": 4437792, + "step": 3415 + }, + { + "epoch": 0.16710233796692156, + "grad_norm": 0.008520863018929958, + "learning_rate": 0.29462438675624114, + "loss": 0.169, + "num_input_tokens_seen": 4443904, + "step": 3420 + }, + { + "epoch": 0.16734663963061588, + "grad_norm": 0.005581583362072706, + "learning_rate": 0.2946087474311506, + "loss": 0.1506, + "num_input_tokens_seen": 4450592, + "step": 3425 + }, + { + "epoch": 0.1675909412943102, + "grad_norm": 0.005354462191462517, + "learning_rate": 0.294593085805574, + "loss": 0.1352, + "num_input_tokens_seen": 4457376, + "step": 3430 + }, + { + "epoch": 0.16783524295800453, + "grad_norm": 0.004989621229469776, + "learning_rate": 0.2945774018819264, + "loss": 0.1496, + "num_input_tokens_seen": 4463392, + "step": 3435 + }, + { + "epoch": 0.16807954462169888, + "grad_norm": 0.014308260753750801, + "learning_rate": 0.2945616956626266, + "loss": 0.1597, + "num_input_tokens_seen": 4469856, + "step": 3440 + }, + { + "epoch": 0.1683238462853932, + "grad_norm": 0.01211550459265709, + "learning_rate": 0.2945459671500966, + "loss": 0.1697, + "num_input_tokens_seen": 4476384, + "step": 3445 + }, + { + "epoch": 0.16856814794908753, + "grad_norm": 0.004750086460262537, + "learning_rate": 0.2945302163467621, + "loss": 0.126, + "num_input_tokens_seen": 4483008, + "step": 3450 + }, + { + "epoch": 0.16881244961278186, + "grad_norm": 0.005010053049772978, + "learning_rate": 0.2945144432550519, + "loss": 0.149, + "num_input_tokens_seen": 4489728, + "step": 3455 + }, + { + "epoch": 0.16905675127647618, + "grad_norm": 0.00448744697496295, + "learning_rate": 0.29449864787739843, + "loss": 0.1438, + "num_input_tokens_seen": 4496320, + "step": 3460 + }, + { + "epoch": 0.16930105294017053, + "grad_norm": 0.006306803785264492, + "learning_rate": 0.2944828302162376, + "loss": 0.1479, + "num_input_tokens_seen": 4502752, + "step": 3465 + }, + { + "epoch": 0.16954535460386486, + "grad_norm": 0.005514460150152445, + "learning_rate": 0.2944669902740087, + "loss": 0.1514, + "num_input_tokens_seen": 4509440, + "step": 3470 + }, + { + "epoch": 0.16978965626755918, + "grad_norm": 0.01155142579227686, + "learning_rate": 0.2944511280531544, + "loss": 0.1408, + "num_input_tokens_seen": 4515968, + "step": 3475 + }, + { + "epoch": 0.1700339579312535, + "grad_norm": 0.005715505685657263, + "learning_rate": 0.29443524355612083, + "loss": 0.1376, + "num_input_tokens_seen": 4522304, + "step": 3480 + }, + { + "epoch": 0.17027825959494783, + "grad_norm": 0.012330758385360241, + "learning_rate": 0.29441933678535764, + "loss": 0.122, + "num_input_tokens_seen": 4528736, + "step": 3485 + }, + { + "epoch": 0.17052256125864218, + "grad_norm": 0.005465715657919645, + "learning_rate": 0.29440340774331786, + "loss": 0.1973, + "num_input_tokens_seen": 4535168, + "step": 3490 + }, + { + "epoch": 0.1707668629223365, + "grad_norm": 0.0036885146982967854, + "learning_rate": 0.2943874564324579, + "loss": 0.1561, + "num_input_tokens_seen": 4541824, + "step": 3495 + }, + { + "epoch": 0.17101116458603083, + "grad_norm": 0.003873786423355341, + "learning_rate": 0.2943714828552376, + "loss": 0.1389, + "num_input_tokens_seen": 4548160, + "step": 3500 + }, + { + "epoch": 0.17125546624972515, + "grad_norm": 0.006759281735867262, + "learning_rate": 0.29435548701412045, + "loss": 0.1455, + "num_input_tokens_seen": 4554496, + "step": 3505 + }, + { + "epoch": 0.1714997679134195, + "grad_norm": 0.005273349117487669, + "learning_rate": 0.2943394689115731, + "loss": 0.1493, + "num_input_tokens_seen": 4560960, + "step": 3510 + }, + { + "epoch": 0.17174406957711383, + "grad_norm": 0.004774583037942648, + "learning_rate": 0.29432342855006577, + "loss": 0.1353, + "num_input_tokens_seen": 4567200, + "step": 3515 + }, + { + "epoch": 0.17198837124080815, + "grad_norm": 0.007701503578573465, + "learning_rate": 0.294307365932072, + "loss": 0.1296, + "num_input_tokens_seen": 4574336, + "step": 3520 + }, + { + "epoch": 0.17223267290450248, + "grad_norm": 0.010509933345019817, + "learning_rate": 0.294291281060069, + "loss": 0.1381, + "num_input_tokens_seen": 4580832, + "step": 3525 + }, + { + "epoch": 0.1724769745681968, + "grad_norm": 0.005069775506854057, + "learning_rate": 0.29427517393653724, + "loss": 0.1624, + "num_input_tokens_seen": 4586752, + "step": 3530 + }, + { + "epoch": 0.17272127623189115, + "grad_norm": 0.0038487182464450598, + "learning_rate": 0.29425904456396046, + "loss": 0.1265, + "num_input_tokens_seen": 4593088, + "step": 3535 + }, + { + "epoch": 0.17296557789558548, + "grad_norm": 0.004181241616606712, + "learning_rate": 0.2942428929448262, + "loss": 0.1432, + "num_input_tokens_seen": 4600128, + "step": 3540 + }, + { + "epoch": 0.1732098795592798, + "grad_norm": 0.007150836754590273, + "learning_rate": 0.2942267190816252, + "loss": 0.1405, + "num_input_tokens_seen": 4606592, + "step": 3545 + }, + { + "epoch": 0.17345418122297412, + "grad_norm": 0.003995053470134735, + "learning_rate": 0.2942105229768516, + "loss": 0.1419, + "num_input_tokens_seen": 4613248, + "step": 3550 + }, + { + "epoch": 0.17369848288666845, + "grad_norm": 0.004407924134284258, + "learning_rate": 0.29419430463300306, + "loss": 0.144, + "num_input_tokens_seen": 4619712, + "step": 3555 + }, + { + "epoch": 0.1739427845503628, + "grad_norm": 0.004128584172576666, + "learning_rate": 0.2941780640525808, + "loss": 0.1299, + "num_input_tokens_seen": 4626016, + "step": 3560 + }, + { + "epoch": 0.17418708621405712, + "grad_norm": 0.007633062079548836, + "learning_rate": 0.2941618012380891, + "loss": 0.1305, + "num_input_tokens_seen": 4632288, + "step": 3565 + }, + { + "epoch": 0.17443138787775145, + "grad_norm": 0.009707351215183735, + "learning_rate": 0.29414551619203605, + "loss": 0.1482, + "num_input_tokens_seen": 4638880, + "step": 3570 + }, + { + "epoch": 0.17467568954144577, + "grad_norm": 0.00554875610396266, + "learning_rate": 0.29412920891693295, + "loss": 0.1752, + "num_input_tokens_seen": 4645056, + "step": 3575 + }, + { + "epoch": 0.1749199912051401, + "grad_norm": 0.00448114238679409, + "learning_rate": 0.2941128794152946, + "loss": 0.1546, + "num_input_tokens_seen": 4651328, + "step": 3580 + }, + { + "epoch": 0.17516429286883445, + "grad_norm": 0.0034846120979636908, + "learning_rate": 0.2940965276896392, + "loss": 0.1604, + "num_input_tokens_seen": 4657888, + "step": 3585 + }, + { + "epoch": 0.17540859453252877, + "grad_norm": 0.002822837559506297, + "learning_rate": 0.2940801537424884, + "loss": 0.1341, + "num_input_tokens_seen": 4664512, + "step": 3590 + }, + { + "epoch": 0.1756528961962231, + "grad_norm": 0.00826732162386179, + "learning_rate": 0.2940637575763673, + "loss": 0.1463, + "num_input_tokens_seen": 4671008, + "step": 3595 + }, + { + "epoch": 0.17589719785991742, + "grad_norm": 0.0058800047263503075, + "learning_rate": 0.2940473391938043, + "loss": 0.1282, + "num_input_tokens_seen": 4677248, + "step": 3600 + }, + { + "epoch": 0.17589719785991742, + "eval_loss": 0.14823652803897858, + "eval_runtime": 402.2713, + "eval_samples_per_second": 90.449, + "eval_steps_per_second": 22.614, + "num_input_tokens_seen": 4677248, + "step": 3600 + }, + { + "epoch": 0.17614149952361174, + "grad_norm": 0.005219560116529465, + "learning_rate": 0.29403089859733145, + "loss": 0.1352, + "num_input_tokens_seen": 4684000, + "step": 3605 + }, + { + "epoch": 0.1763858011873061, + "grad_norm": 0.0047860704362392426, + "learning_rate": 0.294014435789484, + "loss": 0.1638, + "num_input_tokens_seen": 4690336, + "step": 3610 + }, + { + "epoch": 0.17663010285100042, + "grad_norm": 0.01011186558753252, + "learning_rate": 0.2939979507728007, + "loss": 0.1442, + "num_input_tokens_seen": 4696672, + "step": 3615 + }, + { + "epoch": 0.17687440451469474, + "grad_norm": 0.005375186447054148, + "learning_rate": 0.2939814435498239, + "loss": 0.1498, + "num_input_tokens_seen": 4703456, + "step": 3620 + }, + { + "epoch": 0.17711870617838907, + "grad_norm": 0.0025975604075938463, + "learning_rate": 0.29396491412309905, + "loss": 0.1212, + "num_input_tokens_seen": 4710432, + "step": 3625 + }, + { + "epoch": 0.1773630078420834, + "grad_norm": 0.006873182952404022, + "learning_rate": 0.2939483624951753, + "loss": 0.1332, + "num_input_tokens_seen": 4716736, + "step": 3630 + }, + { + "epoch": 0.17760730950577774, + "grad_norm": 0.0047199237160384655, + "learning_rate": 0.2939317886686051, + "loss": 0.1347, + "num_input_tokens_seen": 4723584, + "step": 3635 + }, + { + "epoch": 0.17785161116947207, + "grad_norm": 0.00434064120054245, + "learning_rate": 0.2939151926459443, + "loss": 0.1466, + "num_input_tokens_seen": 4729440, + "step": 3640 + }, + { + "epoch": 0.1780959128331664, + "grad_norm": 0.007025724742561579, + "learning_rate": 0.2938985744297522, + "loss": 0.1445, + "num_input_tokens_seen": 4735648, + "step": 3645 + }, + { + "epoch": 0.17834021449686072, + "grad_norm": 0.006063849665224552, + "learning_rate": 0.29388193402259166, + "loss": 0.1276, + "num_input_tokens_seen": 4741888, + "step": 3650 + }, + { + "epoch": 0.17858451616055504, + "grad_norm": 0.007734061684459448, + "learning_rate": 0.29386527142702873, + "loss": 0.1438, + "num_input_tokens_seen": 4748544, + "step": 3655 + }, + { + "epoch": 0.1788288178242494, + "grad_norm": 0.006008007097989321, + "learning_rate": 0.293848586645633, + "loss": 0.1426, + "num_input_tokens_seen": 4754816, + "step": 3660 + }, + { + "epoch": 0.17907311948794372, + "grad_norm": 0.0037793139927089214, + "learning_rate": 0.2938318796809775, + "loss": 0.1336, + "num_input_tokens_seen": 4761472, + "step": 3665 + }, + { + "epoch": 0.17931742115163804, + "grad_norm": 0.0038399819750338793, + "learning_rate": 0.29381515053563867, + "loss": 0.1477, + "num_input_tokens_seen": 4768064, + "step": 3670 + }, + { + "epoch": 0.17956172281533236, + "grad_norm": 0.005865282844752073, + "learning_rate": 0.29379839921219636, + "loss": 0.1573, + "num_input_tokens_seen": 4774496, + "step": 3675 + }, + { + "epoch": 0.17980602447902672, + "grad_norm": 0.005683218594640493, + "learning_rate": 0.2937816257132338, + "loss": 0.1378, + "num_input_tokens_seen": 4780640, + "step": 3680 + }, + { + "epoch": 0.18005032614272104, + "grad_norm": 0.004016787279397249, + "learning_rate": 0.2937648300413376, + "loss": 0.1531, + "num_input_tokens_seen": 4786656, + "step": 3685 + }, + { + "epoch": 0.18029462780641536, + "grad_norm": 0.005890519358217716, + "learning_rate": 0.293748012199098, + "loss": 0.1541, + "num_input_tokens_seen": 4793504, + "step": 3690 + }, + { + "epoch": 0.1805389294701097, + "grad_norm": 0.006219967734068632, + "learning_rate": 0.29373117218910844, + "loss": 0.1338, + "num_input_tokens_seen": 4799680, + "step": 3695 + }, + { + "epoch": 0.180783231133804, + "grad_norm": 0.005402980372309685, + "learning_rate": 0.2937143100139659, + "loss": 0.1504, + "num_input_tokens_seen": 4806144, + "step": 3700 + }, + { + "epoch": 0.18102753279749836, + "grad_norm": 0.005309350322932005, + "learning_rate": 0.29369742567627083, + "loss": 0.1588, + "num_input_tokens_seen": 4812992, + "step": 3705 + }, + { + "epoch": 0.1812718344611927, + "grad_norm": 0.009750386700034142, + "learning_rate": 0.29368051917862675, + "loss": 0.1554, + "num_input_tokens_seen": 4819232, + "step": 3710 + }, + { + "epoch": 0.181516136124887, + "grad_norm": 0.005296045448631048, + "learning_rate": 0.2936635905236411, + "loss": 0.1226, + "num_input_tokens_seen": 4825568, + "step": 3715 + }, + { + "epoch": 0.18176043778858134, + "grad_norm": 0.0041799466125667095, + "learning_rate": 0.2936466397139244, + "loss": 0.1407, + "num_input_tokens_seen": 4832224, + "step": 3720 + }, + { + "epoch": 0.18200473945227566, + "grad_norm": 0.004095354583114386, + "learning_rate": 0.2936296667520907, + "loss": 0.1307, + "num_input_tokens_seen": 4838144, + "step": 3725 + }, + { + "epoch": 0.18224904111597, + "grad_norm": 0.00765362149104476, + "learning_rate": 0.2936126716407574, + "loss": 0.1368, + "num_input_tokens_seen": 4844320, + "step": 3730 + }, + { + "epoch": 0.18249334277966434, + "grad_norm": 0.0054589007049798965, + "learning_rate": 0.29359565438254537, + "loss": 0.145, + "num_input_tokens_seen": 4851296, + "step": 3735 + }, + { + "epoch": 0.18273764444335866, + "grad_norm": 0.0055501158349215984, + "learning_rate": 0.29357861498007887, + "loss": 0.1339, + "num_input_tokens_seen": 4857536, + "step": 3740 + }, + { + "epoch": 0.18298194610705298, + "grad_norm": 0.003162214532494545, + "learning_rate": 0.29356155343598567, + "loss": 0.1379, + "num_input_tokens_seen": 4863808, + "step": 3745 + }, + { + "epoch": 0.1832262477707473, + "grad_norm": 0.0035257379058748484, + "learning_rate": 0.2935444697528968, + "loss": 0.1294, + "num_input_tokens_seen": 4870080, + "step": 3750 + }, + { + "epoch": 0.18347054943444166, + "grad_norm": 0.0071447161026299, + "learning_rate": 0.2935273639334468, + "loss": 0.1468, + "num_input_tokens_seen": 4876864, + "step": 3755 + }, + { + "epoch": 0.18371485109813598, + "grad_norm": 0.004478020127862692, + "learning_rate": 0.29351023598027365, + "loss": 0.1522, + "num_input_tokens_seen": 4883456, + "step": 3760 + }, + { + "epoch": 0.1839591527618303, + "grad_norm": 0.006183025427162647, + "learning_rate": 0.2934930858960186, + "loss": 0.1344, + "num_input_tokens_seen": 4889216, + "step": 3765 + }, + { + "epoch": 0.18420345442552463, + "grad_norm": 0.006848332472145557, + "learning_rate": 0.29347591368332643, + "loss": 0.162, + "num_input_tokens_seen": 4895616, + "step": 3770 + }, + { + "epoch": 0.18444775608921896, + "grad_norm": 0.006314761005342007, + "learning_rate": 0.2934587193448454, + "loss": 0.1459, + "num_input_tokens_seen": 4902464, + "step": 3775 + }, + { + "epoch": 0.1846920577529133, + "grad_norm": 0.005280583631247282, + "learning_rate": 0.29344150288322696, + "loss": 0.1446, + "num_input_tokens_seen": 4909280, + "step": 3780 + }, + { + "epoch": 0.18493635941660763, + "grad_norm": 0.004736812319606543, + "learning_rate": 0.2934242643011263, + "loss": 0.1374, + "num_input_tokens_seen": 4915520, + "step": 3785 + }, + { + "epoch": 0.18518066108030196, + "grad_norm": 0.003843441605567932, + "learning_rate": 0.2934070036012016, + "loss": 0.1157, + "num_input_tokens_seen": 4921856, + "step": 3790 + }, + { + "epoch": 0.18542496274399628, + "grad_norm": 0.006553660612553358, + "learning_rate": 0.29338972078611475, + "loss": 0.1652, + "num_input_tokens_seen": 4927776, + "step": 3795 + }, + { + "epoch": 0.1856692644076906, + "grad_norm": 0.006035085767507553, + "learning_rate": 0.2933724158585311, + "loss": 0.1609, + "num_input_tokens_seen": 4934080, + "step": 3800 + }, + { + "epoch": 0.1856692644076906, + "eval_loss": 0.142095685005188, + "eval_runtime": 402.1147, + "eval_samples_per_second": 90.484, + "eval_steps_per_second": 22.623, + "num_input_tokens_seen": 4934080, + "step": 3800 + }, + { + "epoch": 0.18591356607138496, + "grad_norm": 0.004159958101809025, + "learning_rate": 0.29335508882111916, + "loss": 0.128, + "num_input_tokens_seen": 4940224, + "step": 3805 + }, + { + "epoch": 0.18615786773507928, + "grad_norm": 0.003927765879780054, + "learning_rate": 0.29333773967655097, + "loss": 0.1616, + "num_input_tokens_seen": 4947136, + "step": 3810 + }, + { + "epoch": 0.1864021693987736, + "grad_norm": 0.007774055935442448, + "learning_rate": 0.2933203684275021, + "loss": 0.1647, + "num_input_tokens_seen": 4953344, + "step": 3815 + }, + { + "epoch": 0.18664647106246793, + "grad_norm": 0.003865953302010894, + "learning_rate": 0.2933029750766513, + "loss": 0.1456, + "num_input_tokens_seen": 4959840, + "step": 3820 + }, + { + "epoch": 0.18689077272616225, + "grad_norm": 0.0037108373362571, + "learning_rate": 0.2932855596266809, + "loss": 0.1458, + "num_input_tokens_seen": 4966208, + "step": 3825 + }, + { + "epoch": 0.1871350743898566, + "grad_norm": 0.005531928036361933, + "learning_rate": 0.2932681220802765, + "loss": 0.1493, + "num_input_tokens_seen": 4972480, + "step": 3830 + }, + { + "epoch": 0.18737937605355093, + "grad_norm": 0.0050368099473416805, + "learning_rate": 0.2932506624401274, + "loss": 0.15, + "num_input_tokens_seen": 4978944, + "step": 3835 + }, + { + "epoch": 0.18762367771724525, + "grad_norm": 0.0027431356720626354, + "learning_rate": 0.29323318070892584, + "loss": 0.1399, + "num_input_tokens_seen": 4985344, + "step": 3840 + }, + { + "epoch": 0.18786797938093958, + "grad_norm": 0.00445999950170517, + "learning_rate": 0.29321567688936784, + "loss": 0.1478, + "num_input_tokens_seen": 4991744, + "step": 3845 + }, + { + "epoch": 0.18811228104463393, + "grad_norm": 0.004785106983035803, + "learning_rate": 0.29319815098415275, + "loss": 0.1679, + "num_input_tokens_seen": 4998112, + "step": 3850 + }, + { + "epoch": 0.18835658270832825, + "grad_norm": 0.0049592722207307816, + "learning_rate": 0.2931806029959832, + "loss": 0.159, + "num_input_tokens_seen": 5004192, + "step": 3855 + }, + { + "epoch": 0.18860088437202258, + "grad_norm": 0.004371722228825092, + "learning_rate": 0.29316303292756535, + "loss": 0.1345, + "num_input_tokens_seen": 5009984, + "step": 3860 + }, + { + "epoch": 0.1888451860357169, + "grad_norm": 0.002544587245211005, + "learning_rate": 0.29314544078160876, + "loss": 0.1247, + "num_input_tokens_seen": 5016512, + "step": 3865 + }, + { + "epoch": 0.18908948769941122, + "grad_norm": 0.006091092247515917, + "learning_rate": 0.2931278265608263, + "loss": 0.1735, + "num_input_tokens_seen": 5022976, + "step": 3870 + }, + { + "epoch": 0.18933378936310558, + "grad_norm": 0.0033927911426872015, + "learning_rate": 0.29311019026793433, + "loss": 0.195, + "num_input_tokens_seen": 5029184, + "step": 3875 + }, + { + "epoch": 0.1895780910267999, + "grad_norm": 0.009415406733751297, + "learning_rate": 0.29309253190565254, + "loss": 0.1747, + "num_input_tokens_seen": 5035616, + "step": 3880 + }, + { + "epoch": 0.18982239269049422, + "grad_norm": 0.002915463875979185, + "learning_rate": 0.2930748514767042, + "loss": 0.1414, + "num_input_tokens_seen": 5042176, + "step": 3885 + }, + { + "epoch": 0.19006669435418855, + "grad_norm": 0.003552937414497137, + "learning_rate": 0.29305714898381574, + "loss": 0.1477, + "num_input_tokens_seen": 5048800, + "step": 3890 + }, + { + "epoch": 0.19031099601788287, + "grad_norm": 0.002986457897350192, + "learning_rate": 0.29303942442971714, + "loss": 0.1528, + "num_input_tokens_seen": 5055328, + "step": 3895 + }, + { + "epoch": 0.19055529768157722, + "grad_norm": 0.003331694519147277, + "learning_rate": 0.2930216778171417, + "loss": 0.1297, + "num_input_tokens_seen": 5061888, + "step": 3900 + }, + { + "epoch": 0.19079959934527155, + "grad_norm": 0.0034073395654559135, + "learning_rate": 0.2930039091488263, + "loss": 0.1518, + "num_input_tokens_seen": 5068768, + "step": 3905 + }, + { + "epoch": 0.19104390100896587, + "grad_norm": 0.003285859012976289, + "learning_rate": 0.29298611842751093, + "loss": 0.1443, + "num_input_tokens_seen": 5075104, + "step": 3910 + }, + { + "epoch": 0.1912882026726602, + "grad_norm": 0.003833592403680086, + "learning_rate": 0.29296830565593923, + "loss": 0.1357, + "num_input_tokens_seen": 5081312, + "step": 3915 + }, + { + "epoch": 0.19153250433635452, + "grad_norm": 0.0062361592426896095, + "learning_rate": 0.2929504708368582, + "loss": 0.1628, + "num_input_tokens_seen": 5088352, + "step": 3920 + }, + { + "epoch": 0.19177680600004887, + "grad_norm": 0.005190989002585411, + "learning_rate": 0.29293261397301806, + "loss": 0.1605, + "num_input_tokens_seen": 5094528, + "step": 3925 + }, + { + "epoch": 0.1920211076637432, + "grad_norm": 0.0026326256338506937, + "learning_rate": 0.29291473506717275, + "loss": 0.119, + "num_input_tokens_seen": 5101408, + "step": 3930 + }, + { + "epoch": 0.19226540932743752, + "grad_norm": 0.0034382676240056753, + "learning_rate": 0.29289683412207923, + "loss": 0.1291, + "num_input_tokens_seen": 5107808, + "step": 3935 + }, + { + "epoch": 0.19250971099113184, + "grad_norm": 0.004861193243414164, + "learning_rate": 0.29287891114049813, + "loss": 0.1457, + "num_input_tokens_seen": 5114464, + "step": 3940 + }, + { + "epoch": 0.19275401265482617, + "grad_norm": 0.002864973619580269, + "learning_rate": 0.29286096612519347, + "loss": 0.1339, + "num_input_tokens_seen": 5120864, + "step": 3945 + }, + { + "epoch": 0.19299831431852052, + "grad_norm": 0.0038920731749385595, + "learning_rate": 0.2928429990789325, + "loss": 0.1252, + "num_input_tokens_seen": 5127520, + "step": 3950 + }, + { + "epoch": 0.19324261598221484, + "grad_norm": 0.007919884286820889, + "learning_rate": 0.29282501000448596, + "loss": 0.1387, + "num_input_tokens_seen": 5134112, + "step": 3955 + }, + { + "epoch": 0.19348691764590917, + "grad_norm": 0.004622836597263813, + "learning_rate": 0.2928069989046281, + "loss": 0.1603, + "num_input_tokens_seen": 5140480, + "step": 3960 + }, + { + "epoch": 0.1937312193096035, + "grad_norm": 0.0032605002634227276, + "learning_rate": 0.2927889657821363, + "loss": 0.1423, + "num_input_tokens_seen": 5146848, + "step": 3965 + }, + { + "epoch": 0.19397552097329782, + "grad_norm": 0.0035541069228202105, + "learning_rate": 0.2927709106397916, + "loss": 0.1535, + "num_input_tokens_seen": 5153760, + "step": 3970 + }, + { + "epoch": 0.19421982263699217, + "grad_norm": 0.003432899247854948, + "learning_rate": 0.29275283348037834, + "loss": 0.1355, + "num_input_tokens_seen": 5159680, + "step": 3975 + }, + { + "epoch": 0.1944641243006865, + "grad_norm": 0.0037247019354254007, + "learning_rate": 0.29273473430668423, + "loss": 0.1509, + "num_input_tokens_seen": 5166176, + "step": 3980 + }, + { + "epoch": 0.19470842596438082, + "grad_norm": 0.004510248079895973, + "learning_rate": 0.2927166131215003, + "loss": 0.1393, + "num_input_tokens_seen": 5172672, + "step": 3985 + }, + { + "epoch": 0.19495272762807514, + "grad_norm": 0.00299235456623137, + "learning_rate": 0.2926984699276212, + "loss": 0.1269, + "num_input_tokens_seen": 5179040, + "step": 3990 + }, + { + "epoch": 0.19519702929176946, + "grad_norm": 0.0038622368592768908, + "learning_rate": 0.29268030472784473, + "loss": 0.1161, + "num_input_tokens_seen": 5185664, + "step": 3995 + }, + { + "epoch": 0.19544133095546382, + "grad_norm": 0.008787636645138264, + "learning_rate": 0.2926621175249723, + "loss": 0.1478, + "num_input_tokens_seen": 5191936, + "step": 4000 + }, + { + "epoch": 0.19544133095546382, + "eval_loss": 0.13819430768489838, + "eval_runtime": 401.7654, + "eval_samples_per_second": 90.563, + "eval_steps_per_second": 22.643, + "num_input_tokens_seen": 5191936, + "step": 4000 + }, + { + "epoch": 0.19568563261915814, + "grad_norm": 0.004920619539916515, + "learning_rate": 0.29264390832180853, + "loss": 0.1411, + "num_input_tokens_seen": 5198176, + "step": 4005 + }, + { + "epoch": 0.19592993428285246, + "grad_norm": 0.006547709461301565, + "learning_rate": 0.29262567712116144, + "loss": 0.1442, + "num_input_tokens_seen": 5204320, + "step": 4010 + }, + { + "epoch": 0.1961742359465468, + "grad_norm": 0.0058131893165409565, + "learning_rate": 0.29260742392584266, + "loss": 0.1346, + "num_input_tokens_seen": 5211232, + "step": 4015 + }, + { + "epoch": 0.19641853761024114, + "grad_norm": 0.005106750410050154, + "learning_rate": 0.292589148738667, + "loss": 0.128, + "num_input_tokens_seen": 5217600, + "step": 4020 + }, + { + "epoch": 0.19666283927393546, + "grad_norm": 0.007514344062656164, + "learning_rate": 0.2925708515624527, + "loss": 0.1332, + "num_input_tokens_seen": 5224544, + "step": 4025 + }, + { + "epoch": 0.1969071409376298, + "grad_norm": 0.003640830749645829, + "learning_rate": 0.29255253240002144, + "loss": 0.1209, + "num_input_tokens_seen": 5231488, + "step": 4030 + }, + { + "epoch": 0.1971514426013241, + "grad_norm": 0.0052973441779613495, + "learning_rate": 0.2925341912541983, + "loss": 0.1331, + "num_input_tokens_seen": 5237760, + "step": 4035 + }, + { + "epoch": 0.19739574426501844, + "grad_norm": 0.006323716137558222, + "learning_rate": 0.2925158281278116, + "loss": 0.1513, + "num_input_tokens_seen": 5244032, + "step": 4040 + }, + { + "epoch": 0.1976400459287128, + "grad_norm": 0.0027939993888139725, + "learning_rate": 0.29249744302369324, + "loss": 0.1376, + "num_input_tokens_seen": 5250080, + "step": 4045 + }, + { + "epoch": 0.1978843475924071, + "grad_norm": 0.003994588274508715, + "learning_rate": 0.29247903594467844, + "loss": 0.1308, + "num_input_tokens_seen": 5256896, + "step": 4050 + }, + { + "epoch": 0.19812864925610144, + "grad_norm": 0.004751657601445913, + "learning_rate": 0.2924606068936058, + "loss": 0.1387, + "num_input_tokens_seen": 5263712, + "step": 4055 + }, + { + "epoch": 0.19837295091979576, + "grad_norm": 0.003407130017876625, + "learning_rate": 0.2924421558733173, + "loss": 0.1296, + "num_input_tokens_seen": 5270048, + "step": 4060 + }, + { + "epoch": 0.19861725258349008, + "grad_norm": 0.01019314769655466, + "learning_rate": 0.2924236828866583, + "loss": 0.16, + "num_input_tokens_seen": 5276448, + "step": 4065 + }, + { + "epoch": 0.19886155424718444, + "grad_norm": 0.004511298146098852, + "learning_rate": 0.29240518793647763, + "loss": 0.1201, + "num_input_tokens_seen": 5282848, + "step": 4070 + }, + { + "epoch": 0.19910585591087876, + "grad_norm": 0.003981330431997776, + "learning_rate": 0.29238667102562743, + "loss": 0.1275, + "num_input_tokens_seen": 5289056, + "step": 4075 + }, + { + "epoch": 0.19935015757457308, + "grad_norm": 0.006935680750757456, + "learning_rate": 0.29236813215696317, + "loss": 0.1324, + "num_input_tokens_seen": 5295328, + "step": 4080 + }, + { + "epoch": 0.1995944592382674, + "grad_norm": 0.006028303410857916, + "learning_rate": 0.2923495713333439, + "loss": 0.1258, + "num_input_tokens_seen": 5301632, + "step": 4085 + }, + { + "epoch": 0.19983876090196173, + "grad_norm": 0.007293918170034885, + "learning_rate": 0.29233098855763173, + "loss": 0.1583, + "num_input_tokens_seen": 5308384, + "step": 4090 + }, + { + "epoch": 0.20008306256565608, + "grad_norm": 0.004393852315843105, + "learning_rate": 0.29231238383269254, + "loss": 0.1374, + "num_input_tokens_seen": 5314688, + "step": 4095 + }, + { + "epoch": 0.2003273642293504, + "grad_norm": 0.005952693521976471, + "learning_rate": 0.2922937571613954, + "loss": 0.1394, + "num_input_tokens_seen": 5320480, + "step": 4100 + }, + { + "epoch": 0.20057166589304473, + "grad_norm": 0.005263230763375759, + "learning_rate": 0.29227510854661265, + "loss": 0.1365, + "num_input_tokens_seen": 5327168, + "step": 4105 + }, + { + "epoch": 0.20081596755673906, + "grad_norm": 0.006189871113747358, + "learning_rate": 0.29225643799122025, + "loss": 0.1365, + "num_input_tokens_seen": 5333920, + "step": 4110 + }, + { + "epoch": 0.20106026922043338, + "grad_norm": 0.0049432567320764065, + "learning_rate": 0.2922377454980974, + "loss": 0.144, + "num_input_tokens_seen": 5340320, + "step": 4115 + }, + { + "epoch": 0.20130457088412773, + "grad_norm": 0.003958342596888542, + "learning_rate": 0.29221903107012676, + "loss": 0.1295, + "num_input_tokens_seen": 5346592, + "step": 4120 + }, + { + "epoch": 0.20154887254782206, + "grad_norm": 0.006187561433762312, + "learning_rate": 0.29220029471019426, + "loss": 0.1408, + "num_input_tokens_seen": 5353504, + "step": 4125 + }, + { + "epoch": 0.20179317421151638, + "grad_norm": 0.00486937677487731, + "learning_rate": 0.2921815364211893, + "loss": 0.134, + "num_input_tokens_seen": 5360096, + "step": 4130 + }, + { + "epoch": 0.2020374758752107, + "grad_norm": 0.005541263148188591, + "learning_rate": 0.29216275620600474, + "loss": 0.1142, + "num_input_tokens_seen": 5366688, + "step": 4135 + }, + { + "epoch": 0.20228177753890503, + "grad_norm": 0.0040527270175516605, + "learning_rate": 0.29214395406753657, + "loss": 0.1397, + "num_input_tokens_seen": 5373120, + "step": 4140 + }, + { + "epoch": 0.20252607920259938, + "grad_norm": 0.0036926267202943563, + "learning_rate": 0.2921251300086844, + "loss": 0.1314, + "num_input_tokens_seen": 5379840, + "step": 4145 + }, + { + "epoch": 0.2027703808662937, + "grad_norm": 0.004110809415578842, + "learning_rate": 0.2921062840323511, + "loss": 0.1465, + "num_input_tokens_seen": 5386464, + "step": 4150 + }, + { + "epoch": 0.20301468252998803, + "grad_norm": 0.006467475555837154, + "learning_rate": 0.29208741614144307, + "loss": 0.1442, + "num_input_tokens_seen": 5392896, + "step": 4155 + }, + { + "epoch": 0.20325898419368235, + "grad_norm": 0.007874083705246449, + "learning_rate": 0.2920685263388698, + "loss": 0.1277, + "num_input_tokens_seen": 5399168, + "step": 4160 + }, + { + "epoch": 0.20350328585737668, + "grad_norm": 0.006010261829942465, + "learning_rate": 0.2920496146275445, + "loss": 0.1375, + "num_input_tokens_seen": 5405472, + "step": 4165 + }, + { + "epoch": 0.20374758752107103, + "grad_norm": 0.006290101911872625, + "learning_rate": 0.29203068101038343, + "loss": 0.1575, + "num_input_tokens_seen": 5411840, + "step": 4170 + }, + { + "epoch": 0.20399188918476535, + "grad_norm": 0.008122577331960201, + "learning_rate": 0.2920117254903065, + "loss": 0.1478, + "num_input_tokens_seen": 5418528, + "step": 4175 + }, + { + "epoch": 0.20423619084845968, + "grad_norm": 0.006215281318873167, + "learning_rate": 0.29199274807023695, + "loss": 0.1704, + "num_input_tokens_seen": 5424832, + "step": 4180 + }, + { + "epoch": 0.204480492512154, + "grad_norm": 0.0037066449876874685, + "learning_rate": 0.29197374875310117, + "loss": 0.1379, + "num_input_tokens_seen": 5431616, + "step": 4185 + }, + { + "epoch": 0.20472479417584835, + "grad_norm": 0.004425386898219585, + "learning_rate": 0.2919547275418292, + "loss": 0.168, + "num_input_tokens_seen": 5438048, + "step": 4190 + }, + { + "epoch": 0.20496909583954268, + "grad_norm": 0.005485829897224903, + "learning_rate": 0.29193568443935436, + "loss": 0.1377, + "num_input_tokens_seen": 5444512, + "step": 4195 + }, + { + "epoch": 0.205213397503237, + "grad_norm": 0.00548839196562767, + "learning_rate": 0.2919166194486133, + "loss": 0.1375, + "num_input_tokens_seen": 5451200, + "step": 4200 + }, + { + "epoch": 0.205213397503237, + "eval_loss": 0.1395796537399292, + "eval_runtime": 402.1895, + "eval_samples_per_second": 90.467, + "eval_steps_per_second": 22.619, + "num_input_tokens_seen": 5451200, + "step": 4200 + }, + { + "epoch": 0.20545769916693132, + "grad_norm": 0.006534198764711618, + "learning_rate": 0.2918975325725461, + "loss": 0.1418, + "num_input_tokens_seen": 5457888, + "step": 4205 + }, + { + "epoch": 0.20570200083062565, + "grad_norm": 0.004309219773858786, + "learning_rate": 0.29187842381409607, + "loss": 0.1595, + "num_input_tokens_seen": 5464256, + "step": 4210 + }, + { + "epoch": 0.20594630249432, + "grad_norm": 0.004149230197072029, + "learning_rate": 0.29185929317621023, + "loss": 0.1458, + "num_input_tokens_seen": 5471392, + "step": 4215 + }, + { + "epoch": 0.20619060415801432, + "grad_norm": 0.00408179173246026, + "learning_rate": 0.29184014066183867, + "loss": 0.1486, + "num_input_tokens_seen": 5477888, + "step": 4220 + }, + { + "epoch": 0.20643490582170865, + "grad_norm": 0.005921475123614073, + "learning_rate": 0.2918209662739349, + "loss": 0.1286, + "num_input_tokens_seen": 5484416, + "step": 4225 + }, + { + "epoch": 0.20667920748540297, + "grad_norm": 0.008027811534702778, + "learning_rate": 0.29180177001545593, + "loss": 0.1555, + "num_input_tokens_seen": 5491136, + "step": 4230 + }, + { + "epoch": 0.2069235091490973, + "grad_norm": 0.004586419556289911, + "learning_rate": 0.29178255188936203, + "loss": 0.1641, + "num_input_tokens_seen": 5497280, + "step": 4235 + }, + { + "epoch": 0.20716781081279165, + "grad_norm": 0.004143036901950836, + "learning_rate": 0.2917633118986169, + "loss": 0.1289, + "num_input_tokens_seen": 5504128, + "step": 4240 + }, + { + "epoch": 0.20741211247648597, + "grad_norm": 0.0057370332069695, + "learning_rate": 0.2917440500461875, + "loss": 0.1842, + "num_input_tokens_seen": 5511872, + "step": 4245 + }, + { + "epoch": 0.2076564141401803, + "grad_norm": 0.004800648428499699, + "learning_rate": 0.29172476633504435, + "loss": 0.1421, + "num_input_tokens_seen": 5518336, + "step": 4250 + }, + { + "epoch": 0.20790071580387462, + "grad_norm": 0.006702306214720011, + "learning_rate": 0.2917054607681612, + "loss": 0.1374, + "num_input_tokens_seen": 5524384, + "step": 4255 + }, + { + "epoch": 0.20814501746756894, + "grad_norm": 0.006671188399195671, + "learning_rate": 0.29168613334851523, + "loss": 0.181, + "num_input_tokens_seen": 5530752, + "step": 4260 + }, + { + "epoch": 0.2083893191312633, + "grad_norm": 0.003431736258789897, + "learning_rate": 0.2916667840790869, + "loss": 0.1649, + "num_input_tokens_seen": 5537408, + "step": 4265 + }, + { + "epoch": 0.20863362079495762, + "grad_norm": 0.004184196703135967, + "learning_rate": 0.2916474129628603, + "loss": 0.1333, + "num_input_tokens_seen": 5543776, + "step": 4270 + }, + { + "epoch": 0.20887792245865194, + "grad_norm": 0.0038570330943912268, + "learning_rate": 0.29162802000282245, + "loss": 0.1293, + "num_input_tokens_seen": 5550208, + "step": 4275 + }, + { + "epoch": 0.20912222412234627, + "grad_norm": 0.0032389082480221987, + "learning_rate": 0.2916086052019642, + "loss": 0.1346, + "num_input_tokens_seen": 5556736, + "step": 4280 + }, + { + "epoch": 0.2093665257860406, + "grad_norm": 0.004164968151599169, + "learning_rate": 0.2915891685632794, + "loss": 0.1589, + "num_input_tokens_seen": 5563136, + "step": 4285 + }, + { + "epoch": 0.20961082744973494, + "grad_norm": 0.0029264141339808702, + "learning_rate": 0.29156971008976545, + "loss": 0.119, + "num_input_tokens_seen": 5569824, + "step": 4290 + }, + { + "epoch": 0.20985512911342927, + "grad_norm": 0.004589404910802841, + "learning_rate": 0.2915502297844232, + "loss": 0.152, + "num_input_tokens_seen": 5576384, + "step": 4295 + }, + { + "epoch": 0.2100994307771236, + "grad_norm": 0.009353786706924438, + "learning_rate": 0.2915307276502566, + "loss": 0.169, + "num_input_tokens_seen": 5582656, + "step": 4300 + }, + { + "epoch": 0.21034373244081792, + "grad_norm": 0.007468708325177431, + "learning_rate": 0.29151120369027334, + "loss": 0.1488, + "num_input_tokens_seen": 5589312, + "step": 4305 + }, + { + "epoch": 0.21058803410451224, + "grad_norm": 0.002848197240382433, + "learning_rate": 0.29149165790748405, + "loss": 0.1309, + "num_input_tokens_seen": 5596000, + "step": 4310 + }, + { + "epoch": 0.2108323357682066, + "grad_norm": 0.003932049032300711, + "learning_rate": 0.291472090304903, + "loss": 0.1666, + "num_input_tokens_seen": 5602528, + "step": 4315 + }, + { + "epoch": 0.21107663743190092, + "grad_norm": 0.0035994071513414383, + "learning_rate": 0.2914525008855478, + "loss": 0.1273, + "num_input_tokens_seen": 5609184, + "step": 4320 + }, + { + "epoch": 0.21132093909559524, + "grad_norm": 0.003262303536757827, + "learning_rate": 0.2914328896524394, + "loss": 0.1393, + "num_input_tokens_seen": 5615648, + "step": 4325 + }, + { + "epoch": 0.21156524075928956, + "grad_norm": 0.005344129633158445, + "learning_rate": 0.291413256608602, + "loss": 0.1499, + "num_input_tokens_seen": 5622432, + "step": 4330 + }, + { + "epoch": 0.2118095424229839, + "grad_norm": 0.008335319347679615, + "learning_rate": 0.29139360175706336, + "loss": 0.1644, + "num_input_tokens_seen": 5628832, + "step": 4335 + }, + { + "epoch": 0.21205384408667824, + "grad_norm": 0.003774066222831607, + "learning_rate": 0.2913739251008544, + "loss": 0.1683, + "num_input_tokens_seen": 5634912, + "step": 4340 + }, + { + "epoch": 0.21229814575037256, + "grad_norm": 0.008106192573904991, + "learning_rate": 0.29135422664300964, + "loss": 0.138, + "num_input_tokens_seen": 5641184, + "step": 4345 + }, + { + "epoch": 0.2125424474140669, + "grad_norm": 0.003908166661858559, + "learning_rate": 0.29133450638656677, + "loss": 0.1523, + "num_input_tokens_seen": 5647648, + "step": 4350 + }, + { + "epoch": 0.2127867490777612, + "grad_norm": 0.00907465536147356, + "learning_rate": 0.2913147643345669, + "loss": 0.1649, + "num_input_tokens_seen": 5653920, + "step": 4355 + }, + { + "epoch": 0.21303105074145554, + "grad_norm": 0.0036871707998216152, + "learning_rate": 0.29129500049005447, + "loss": 0.135, + "num_input_tokens_seen": 5660320, + "step": 4360 + }, + { + "epoch": 0.2132753524051499, + "grad_norm": 0.0043722582049667835, + "learning_rate": 0.2912752148560773, + "loss": 0.1552, + "num_input_tokens_seen": 5666464, + "step": 4365 + }, + { + "epoch": 0.2135196540688442, + "grad_norm": 0.0038825483061373234, + "learning_rate": 0.2912554074356866, + "loss": 0.1309, + "num_input_tokens_seen": 5672992, + "step": 4370 + }, + { + "epoch": 0.21376395573253854, + "grad_norm": 0.004383559804409742, + "learning_rate": 0.2912355782319371, + "loss": 0.1371, + "num_input_tokens_seen": 5679520, + "step": 4375 + }, + { + "epoch": 0.21400825739623286, + "grad_norm": 0.00546663161367178, + "learning_rate": 0.2912157272478864, + "loss": 0.131, + "num_input_tokens_seen": 5685696, + "step": 4380 + }, + { + "epoch": 0.2142525590599272, + "grad_norm": 0.003911165986210108, + "learning_rate": 0.291195854486596, + "loss": 0.1325, + "num_input_tokens_seen": 5691968, + "step": 4385 + }, + { + "epoch": 0.21449686072362154, + "grad_norm": 0.0033889079932123423, + "learning_rate": 0.2911759599511305, + "loss": 0.1614, + "num_input_tokens_seen": 5698144, + "step": 4390 + }, + { + "epoch": 0.21474116238731586, + "grad_norm": 0.004077570978552103, + "learning_rate": 0.29115604364455777, + "loss": 0.1523, + "num_input_tokens_seen": 5704800, + "step": 4395 + }, + { + "epoch": 0.21498546405101018, + "grad_norm": 0.0029229833744466305, + "learning_rate": 0.2911361055699493, + "loss": 0.1283, + "num_input_tokens_seen": 5711648, + "step": 4400 + }, + { + "epoch": 0.21498546405101018, + "eval_loss": 0.1436413824558258, + "eval_runtime": 402.0539, + "eval_samples_per_second": 90.498, + "eval_steps_per_second": 22.626, + "num_input_tokens_seen": 5711648, + "step": 4400 + }, + { + "epoch": 0.2152297657147045, + "grad_norm": 0.0029290823731571436, + "learning_rate": 0.2911161457303797, + "loss": 0.1477, + "num_input_tokens_seen": 5718112, + "step": 4405 + }, + { + "epoch": 0.21547406737839886, + "grad_norm": 0.0028894494753330946, + "learning_rate": 0.291096164128927, + "loss": 0.1366, + "num_input_tokens_seen": 5724576, + "step": 4410 + }, + { + "epoch": 0.21571836904209318, + "grad_norm": 0.004369703121483326, + "learning_rate": 0.2910761607686727, + "loss": 0.1538, + "num_input_tokens_seen": 5730880, + "step": 4415 + }, + { + "epoch": 0.2159626707057875, + "grad_norm": 0.003256073221564293, + "learning_rate": 0.2910561356527016, + "loss": 0.1518, + "num_input_tokens_seen": 5737568, + "step": 4420 + }, + { + "epoch": 0.21620697236948183, + "grad_norm": 0.008569875732064247, + "learning_rate": 0.2910360887841017, + "loss": 0.1949, + "num_input_tokens_seen": 5744000, + "step": 4425 + }, + { + "epoch": 0.21645127403317616, + "grad_norm": 0.0037908265367150307, + "learning_rate": 0.2910160201659645, + "loss": 0.1455, + "num_input_tokens_seen": 5750560, + "step": 4430 + }, + { + "epoch": 0.2166955756968705, + "grad_norm": 0.0037245475687086582, + "learning_rate": 0.29099592980138494, + "loss": 0.1332, + "num_input_tokens_seen": 5756800, + "step": 4435 + }, + { + "epoch": 0.21693987736056483, + "grad_norm": 0.0032193236984312534, + "learning_rate": 0.29097581769346115, + "loss": 0.1438, + "num_input_tokens_seen": 5762976, + "step": 4440 + }, + { + "epoch": 0.21718417902425916, + "grad_norm": 0.0032669599168002605, + "learning_rate": 0.29095568384529463, + "loss": 0.153, + "num_input_tokens_seen": 5770176, + "step": 4445 + }, + { + "epoch": 0.21742848068795348, + "grad_norm": 0.005371781066060066, + "learning_rate": 0.2909355282599903, + "loss": 0.1586, + "num_input_tokens_seen": 5776512, + "step": 4450 + }, + { + "epoch": 0.2176727823516478, + "grad_norm": 0.002377155004069209, + "learning_rate": 0.29091535094065635, + "loss": 0.1337, + "num_input_tokens_seen": 5782720, + "step": 4455 + }, + { + "epoch": 0.21791708401534216, + "grad_norm": 0.003807287896052003, + "learning_rate": 0.2908951518904045, + "loss": 0.1404, + "num_input_tokens_seen": 5788736, + "step": 4460 + }, + { + "epoch": 0.21816138567903648, + "grad_norm": 0.0032387799583375454, + "learning_rate": 0.29087493111234963, + "loss": 0.1539, + "num_input_tokens_seen": 5795328, + "step": 4465 + }, + { + "epoch": 0.2184056873427308, + "grad_norm": 0.003639101516455412, + "learning_rate": 0.29085468860961, + "loss": 0.1235, + "num_input_tokens_seen": 5801344, + "step": 4470 + }, + { + "epoch": 0.21864998900642513, + "grad_norm": 0.004084721673280001, + "learning_rate": 0.2908344243853073, + "loss": 0.1504, + "num_input_tokens_seen": 5807776, + "step": 4475 + }, + { + "epoch": 0.21889429067011945, + "grad_norm": 0.0028435245621949434, + "learning_rate": 0.2908141384425666, + "loss": 0.1465, + "num_input_tokens_seen": 5814304, + "step": 4480 + }, + { + "epoch": 0.2191385923338138, + "grad_norm": 0.0039061401039361954, + "learning_rate": 0.2907938307845161, + "loss": 0.1182, + "num_input_tokens_seen": 5820992, + "step": 4485 + }, + { + "epoch": 0.21938289399750813, + "grad_norm": 0.00815920066088438, + "learning_rate": 0.2907735014142876, + "loss": 0.1617, + "num_input_tokens_seen": 5827840, + "step": 4490 + }, + { + "epoch": 0.21962719566120245, + "grad_norm": 0.004781123250722885, + "learning_rate": 0.2907531503350161, + "loss": 0.1268, + "num_input_tokens_seen": 5833920, + "step": 4495 + }, + { + "epoch": 0.21987149732489678, + "grad_norm": 0.0034679702948778868, + "learning_rate": 0.29073277754983995, + "loss": 0.1503, + "num_input_tokens_seen": 5840064, + "step": 4500 + }, + { + "epoch": 0.2201157989885911, + "grad_norm": 0.0030624668579548597, + "learning_rate": 0.290712383061901, + "loss": 0.1293, + "num_input_tokens_seen": 5846400, + "step": 4505 + }, + { + "epoch": 0.22036010065228545, + "grad_norm": 0.002656269120052457, + "learning_rate": 0.2906919668743443, + "loss": 0.1573, + "num_input_tokens_seen": 5852544, + "step": 4510 + }, + { + "epoch": 0.22060440231597978, + "grad_norm": 0.002597316401079297, + "learning_rate": 0.29067152899031823, + "loss": 0.1581, + "num_input_tokens_seen": 5858464, + "step": 4515 + }, + { + "epoch": 0.2208487039796741, + "grad_norm": 0.0022498099133372307, + "learning_rate": 0.2906510694129746, + "loss": 0.1479, + "num_input_tokens_seen": 5864864, + "step": 4520 + }, + { + "epoch": 0.22109300564336842, + "grad_norm": 0.0025856096763163805, + "learning_rate": 0.2906305881454685, + "loss": 0.1202, + "num_input_tokens_seen": 5871552, + "step": 4525 + }, + { + "epoch": 0.22133730730706275, + "grad_norm": 0.002537765773013234, + "learning_rate": 0.2906100851909585, + "loss": 0.136, + "num_input_tokens_seen": 5877888, + "step": 4530 + }, + { + "epoch": 0.2215816089707571, + "grad_norm": 0.004915823228657246, + "learning_rate": 0.29058956055260626, + "loss": 0.1386, + "num_input_tokens_seen": 5884416, + "step": 4535 + }, + { + "epoch": 0.22182591063445142, + "grad_norm": 0.003281382843852043, + "learning_rate": 0.2905690142335771, + "loss": 0.1467, + "num_input_tokens_seen": 5891456, + "step": 4540 + }, + { + "epoch": 0.22207021229814575, + "grad_norm": 0.0033787053544074297, + "learning_rate": 0.29054844623703946, + "loss": 0.1323, + "num_input_tokens_seen": 5897728, + "step": 4545 + }, + { + "epoch": 0.22231451396184007, + "grad_norm": 0.006140438374131918, + "learning_rate": 0.2905278565661651, + "loss": 0.138, + "num_input_tokens_seen": 5904640, + "step": 4550 + }, + { + "epoch": 0.22255881562553442, + "grad_norm": 0.006658319849520922, + "learning_rate": 0.2905072452241293, + "loss": 0.135, + "num_input_tokens_seen": 5910976, + "step": 4555 + }, + { + "epoch": 0.22280311728922875, + "grad_norm": 0.0042272028513252735, + "learning_rate": 0.2904866122141106, + "loss": 0.1428, + "num_input_tokens_seen": 5917248, + "step": 4560 + }, + { + "epoch": 0.22304741895292307, + "grad_norm": 0.003121185814961791, + "learning_rate": 0.2904659575392908, + "loss": 0.1257, + "num_input_tokens_seen": 5923520, + "step": 4565 + }, + { + "epoch": 0.2232917206166174, + "grad_norm": 0.0033863738644868135, + "learning_rate": 0.2904452812028551, + "loss": 0.1313, + "num_input_tokens_seen": 5929792, + "step": 4570 + }, + { + "epoch": 0.22353602228031172, + "grad_norm": 0.00487141078338027, + "learning_rate": 0.2904245832079922, + "loss": 0.1515, + "num_input_tokens_seen": 5936416, + "step": 4575 + }, + { + "epoch": 0.22378032394400607, + "grad_norm": 0.003733778605237603, + "learning_rate": 0.29040386355789377, + "loss": 0.1487, + "num_input_tokens_seen": 5943360, + "step": 4580 + }, + { + "epoch": 0.2240246256077004, + "grad_norm": 0.004387166351079941, + "learning_rate": 0.29038312225575524, + "loss": 0.1256, + "num_input_tokens_seen": 5950080, + "step": 4585 + }, + { + "epoch": 0.22426892727139472, + "grad_norm": 0.004549412522464991, + "learning_rate": 0.29036235930477505, + "loss": 0.1267, + "num_input_tokens_seen": 5957280, + "step": 4590 + }, + { + "epoch": 0.22451322893508904, + "grad_norm": 0.003586435690522194, + "learning_rate": 0.29034157470815514, + "loss": 0.1354, + "num_input_tokens_seen": 5963712, + "step": 4595 + }, + { + "epoch": 0.22475753059878337, + "grad_norm": 0.005158006679266691, + "learning_rate": 0.2903207684691008, + "loss": 0.1314, + "num_input_tokens_seen": 5970048, + "step": 4600 + }, + { + "epoch": 0.22475753059878337, + "eval_loss": 0.13764089345932007, + "eval_runtime": 401.9623, + "eval_samples_per_second": 90.518, + "eval_steps_per_second": 22.631, + "num_input_tokens_seen": 5970048, + "step": 4600 + }, + { + "epoch": 0.22500183226247772, + "grad_norm": 0.003115572966635227, + "learning_rate": 0.29029994059082054, + "loss": 0.1382, + "num_input_tokens_seen": 5976416, + "step": 4605 + }, + { + "epoch": 0.22524613392617204, + "grad_norm": 0.004205652512609959, + "learning_rate": 0.2902790910765264, + "loss": 0.1171, + "num_input_tokens_seen": 5982976, + "step": 4610 + }, + { + "epoch": 0.22549043558986637, + "grad_norm": 0.005134414415806532, + "learning_rate": 0.29025821992943346, + "loss": 0.152, + "num_input_tokens_seen": 5989152, + "step": 4615 + }, + { + "epoch": 0.2257347372535607, + "grad_norm": 0.004132261034101248, + "learning_rate": 0.29023732715276046, + "loss": 0.1239, + "num_input_tokens_seen": 5995552, + "step": 4620 + }, + { + "epoch": 0.22597903891725502, + "grad_norm": 0.0024402861017733812, + "learning_rate": 0.2902164127497293, + "loss": 0.1344, + "num_input_tokens_seen": 6001824, + "step": 4625 + }, + { + "epoch": 0.22622334058094937, + "grad_norm": 0.004677369259297848, + "learning_rate": 0.2901954767235652, + "loss": 0.1217, + "num_input_tokens_seen": 6007936, + "step": 4630 + }, + { + "epoch": 0.2264676422446437, + "grad_norm": 0.003997212741523981, + "learning_rate": 0.2901745190774968, + "loss": 0.1344, + "num_input_tokens_seen": 6014496, + "step": 4635 + }, + { + "epoch": 0.22671194390833802, + "grad_norm": 0.0036205295473337173, + "learning_rate": 0.290153539814756, + "loss": 0.1706, + "num_input_tokens_seen": 6020928, + "step": 4640 + }, + { + "epoch": 0.22695624557203234, + "grad_norm": 0.0039353277534246445, + "learning_rate": 0.2901325389385781, + "loss": 0.136, + "num_input_tokens_seen": 6027456, + "step": 4645 + }, + { + "epoch": 0.22720054723572666, + "grad_norm": 0.002605627989396453, + "learning_rate": 0.2901115164522016, + "loss": 0.1405, + "num_input_tokens_seen": 6033632, + "step": 4650 + }, + { + "epoch": 0.22744484889942101, + "grad_norm": 0.003144359914585948, + "learning_rate": 0.29009047235886865, + "loss": 0.1611, + "num_input_tokens_seen": 6039936, + "step": 4655 + }, + { + "epoch": 0.22768915056311534, + "grad_norm": 0.004494964610785246, + "learning_rate": 0.2900694066618243, + "loss": 0.1417, + "num_input_tokens_seen": 6046848, + "step": 4660 + }, + { + "epoch": 0.22793345222680966, + "grad_norm": 0.00372843979857862, + "learning_rate": 0.2900483193643172, + "loss": 0.1288, + "num_input_tokens_seen": 6052736, + "step": 4665 + }, + { + "epoch": 0.228177753890504, + "grad_norm": 0.0035504845436662436, + "learning_rate": 0.29002721046959934, + "loss": 0.142, + "num_input_tokens_seen": 6059008, + "step": 4670 + }, + { + "epoch": 0.2284220555541983, + "grad_norm": 0.004115863237529993, + "learning_rate": 0.29000607998092587, + "loss": 0.1155, + "num_input_tokens_seen": 6065536, + "step": 4675 + }, + { + "epoch": 0.22866635721789266, + "grad_norm": 0.008130514994263649, + "learning_rate": 0.2899849279015555, + "loss": 0.1433, + "num_input_tokens_seen": 6072288, + "step": 4680 + }, + { + "epoch": 0.228910658881587, + "grad_norm": 0.005149946082383394, + "learning_rate": 0.28996375423475007, + "loss": 0.1609, + "num_input_tokens_seen": 6078560, + "step": 4685 + }, + { + "epoch": 0.2291549605452813, + "grad_norm": 0.003784746862947941, + "learning_rate": 0.28994255898377486, + "loss": 0.1326, + "num_input_tokens_seen": 6084960, + "step": 4690 + }, + { + "epoch": 0.22939926220897564, + "grad_norm": 0.0037589012645184994, + "learning_rate": 0.2899213421518984, + "loss": 0.1137, + "num_input_tokens_seen": 6091392, + "step": 4695 + }, + { + "epoch": 0.22964356387266996, + "grad_norm": 0.004977487493306398, + "learning_rate": 0.2899001037423926, + "loss": 0.1482, + "num_input_tokens_seen": 6097728, + "step": 4700 + }, + { + "epoch": 0.2298878655363643, + "grad_norm": 0.006066678557544947, + "learning_rate": 0.28987884375853273, + "loss": 0.1529, + "num_input_tokens_seen": 6104896, + "step": 4705 + }, + { + "epoch": 0.23013216720005863, + "grad_norm": 0.002540591172873974, + "learning_rate": 0.2898575622035974, + "loss": 0.125, + "num_input_tokens_seen": 6111456, + "step": 4710 + }, + { + "epoch": 0.23037646886375296, + "grad_norm": 0.0057541560381650925, + "learning_rate": 0.2898362590808683, + "loss": 0.1449, + "num_input_tokens_seen": 6117664, + "step": 4715 + }, + { + "epoch": 0.23062077052744728, + "grad_norm": 0.0038955274503678083, + "learning_rate": 0.2898149343936308, + "loss": 0.1336, + "num_input_tokens_seen": 6124224, + "step": 4720 + }, + { + "epoch": 0.23086507219114163, + "grad_norm": 0.0040641063824296, + "learning_rate": 0.2897935881451734, + "loss": 0.136, + "num_input_tokens_seen": 6130720, + "step": 4725 + }, + { + "epoch": 0.23110937385483596, + "grad_norm": 0.0035288978833705187, + "learning_rate": 0.28977222033878797, + "loss": 0.1358, + "num_input_tokens_seen": 6136992, + "step": 4730 + }, + { + "epoch": 0.23135367551853028, + "grad_norm": 0.004429074004292488, + "learning_rate": 0.28975083097776966, + "loss": 0.1331, + "num_input_tokens_seen": 6143264, + "step": 4735 + }, + { + "epoch": 0.2315979771822246, + "grad_norm": 0.0055353082716465, + "learning_rate": 0.28972942006541696, + "loss": 0.1519, + "num_input_tokens_seen": 6149792, + "step": 4740 + }, + { + "epoch": 0.23184227884591893, + "grad_norm": 0.0034338997211307287, + "learning_rate": 0.2897079876050318, + "loss": 0.1382, + "num_input_tokens_seen": 6156160, + "step": 4745 + }, + { + "epoch": 0.23208658050961328, + "grad_norm": 0.0037128457333892584, + "learning_rate": 0.2896865335999192, + "loss": 0.1315, + "num_input_tokens_seen": 6162336, + "step": 4750 + }, + { + "epoch": 0.2323308821733076, + "grad_norm": 0.002697709482163191, + "learning_rate": 0.28966505805338777, + "loss": 0.0962, + "num_input_tokens_seen": 6168896, + "step": 4755 + }, + { + "epoch": 0.23257518383700193, + "grad_norm": 0.005931569263339043, + "learning_rate": 0.2896435609687492, + "loss": 0.1511, + "num_input_tokens_seen": 6175200, + "step": 4760 + }, + { + "epoch": 0.23281948550069625, + "grad_norm": 0.0029333888087421656, + "learning_rate": 0.2896220423493187, + "loss": 0.1144, + "num_input_tokens_seen": 6181664, + "step": 4765 + }, + { + "epoch": 0.23306378716439058, + "grad_norm": 0.005065467674285173, + "learning_rate": 0.28960050219841466, + "loss": 0.1484, + "num_input_tokens_seen": 6188032, + "step": 4770 + }, + { + "epoch": 0.23330808882808493, + "grad_norm": 0.0037458394654095173, + "learning_rate": 0.28957894051935884, + "loss": 0.1095, + "num_input_tokens_seen": 6194560, + "step": 4775 + }, + { + "epoch": 0.23355239049177925, + "grad_norm": 0.004539848770946264, + "learning_rate": 0.2895573573154764, + "loss": 0.1375, + "num_input_tokens_seen": 6201024, + "step": 4780 + }, + { + "epoch": 0.23379669215547358, + "grad_norm": 0.004385912790894508, + "learning_rate": 0.28953575259009556, + "loss": 0.1256, + "num_input_tokens_seen": 6207392, + "step": 4785 + }, + { + "epoch": 0.2340409938191679, + "grad_norm": 0.005787061061710119, + "learning_rate": 0.2895141263465482, + "loss": 0.1459, + "num_input_tokens_seen": 6213792, + "step": 4790 + }, + { + "epoch": 0.23428529548286223, + "grad_norm": 0.0033092680387198925, + "learning_rate": 0.28949247858816934, + "loss": 0.1126, + "num_input_tokens_seen": 6220320, + "step": 4795 + }, + { + "epoch": 0.23452959714655658, + "grad_norm": 0.003867699531838298, + "learning_rate": 0.2894708093182973, + "loss": 0.1196, + "num_input_tokens_seen": 6226272, + "step": 4800 + }, + { + "epoch": 0.23452959714655658, + "eval_loss": 0.1406961977481842, + "eval_runtime": 402.4483, + "eval_samples_per_second": 90.409, + "eval_steps_per_second": 22.604, + "num_input_tokens_seen": 6226272, + "step": 4800 + }, + { + "epoch": 0.2347738988102509, + "grad_norm": 0.006409327033907175, + "learning_rate": 0.2894491185402737, + "loss": 0.1256, + "num_input_tokens_seen": 6232832, + "step": 4805 + }, + { + "epoch": 0.23501820047394523, + "grad_norm": 0.005595401860773563, + "learning_rate": 0.2894274062574437, + "loss": 0.156, + "num_input_tokens_seen": 6239072, + "step": 4810 + }, + { + "epoch": 0.23526250213763955, + "grad_norm": 0.0045680818147957325, + "learning_rate": 0.2894056724731554, + "loss": 0.1311, + "num_input_tokens_seen": 6245344, + "step": 4815 + }, + { + "epoch": 0.23550680380133387, + "grad_norm": 0.003172369906678796, + "learning_rate": 0.28938391719076056, + "loss": 0.1202, + "num_input_tokens_seen": 6251648, + "step": 4820 + }, + { + "epoch": 0.23575110546502823, + "grad_norm": 0.009710987098515034, + "learning_rate": 0.28936214041361413, + "loss": 0.1525, + "num_input_tokens_seen": 6258048, + "step": 4825 + }, + { + "epoch": 0.23599540712872255, + "grad_norm": 0.010787338949739933, + "learning_rate": 0.2893403421450743, + "loss": 0.1538, + "num_input_tokens_seen": 6264448, + "step": 4830 + }, + { + "epoch": 0.23623970879241687, + "grad_norm": 0.0032449448481202126, + "learning_rate": 0.2893185223885026, + "loss": 0.134, + "num_input_tokens_seen": 6271296, + "step": 4835 + }, + { + "epoch": 0.2364840104561112, + "grad_norm": 0.0031164181418716908, + "learning_rate": 0.289296681147264, + "loss": 0.1323, + "num_input_tokens_seen": 6277888, + "step": 4840 + }, + { + "epoch": 0.23672831211980552, + "grad_norm": 0.002808895893394947, + "learning_rate": 0.28927481842472663, + "loss": 0.1544, + "num_input_tokens_seen": 6284416, + "step": 4845 + }, + { + "epoch": 0.23697261378349987, + "grad_norm": 0.003823480335995555, + "learning_rate": 0.28925293422426207, + "loss": 0.1344, + "num_input_tokens_seen": 6290976, + "step": 4850 + }, + { + "epoch": 0.2372169154471942, + "grad_norm": 0.0030135682318359613, + "learning_rate": 0.28923102854924504, + "loss": 0.1381, + "num_input_tokens_seen": 6297248, + "step": 4855 + }, + { + "epoch": 0.23746121711088852, + "grad_norm": 0.00448997039347887, + "learning_rate": 0.2892091014030537, + "loss": 0.139, + "num_input_tokens_seen": 6303968, + "step": 4860 + }, + { + "epoch": 0.23770551877458285, + "grad_norm": 0.005355511792004108, + "learning_rate": 0.2891871527890696, + "loss": 0.1458, + "num_input_tokens_seen": 6310400, + "step": 4865 + }, + { + "epoch": 0.23794982043827717, + "grad_norm": 0.0040593575686216354, + "learning_rate": 0.2891651827106773, + "loss": 0.1426, + "num_input_tokens_seen": 6316864, + "step": 4870 + }, + { + "epoch": 0.23819412210197152, + "grad_norm": 0.00439917529001832, + "learning_rate": 0.2891431911712651, + "loss": 0.1151, + "num_input_tokens_seen": 6323744, + "step": 4875 + }, + { + "epoch": 0.23843842376566585, + "grad_norm": 0.00403089914470911, + "learning_rate": 0.2891211781742241, + "loss": 0.136, + "num_input_tokens_seen": 6330272, + "step": 4880 + }, + { + "epoch": 0.23868272542936017, + "grad_norm": 0.003431793302297592, + "learning_rate": 0.2890991437229492, + "loss": 0.1261, + "num_input_tokens_seen": 6336800, + "step": 4885 + }, + { + "epoch": 0.2389270270930545, + "grad_norm": 0.0052152713760733604, + "learning_rate": 0.2890770878208383, + "loss": 0.1434, + "num_input_tokens_seen": 6343008, + "step": 4890 + }, + { + "epoch": 0.23917132875674885, + "grad_norm": 0.006440270692110062, + "learning_rate": 0.28905501047129273, + "loss": 0.1229, + "num_input_tokens_seen": 6349280, + "step": 4895 + }, + { + "epoch": 0.23941563042044317, + "grad_norm": 0.00581289641559124, + "learning_rate": 0.289032911677717, + "loss": 0.1501, + "num_input_tokens_seen": 6355808, + "step": 4900 + }, + { + "epoch": 0.2396599320841375, + "grad_norm": 0.002963148755952716, + "learning_rate": 0.28901079144351915, + "loss": 0.1344, + "num_input_tokens_seen": 6362208, + "step": 4905 + }, + { + "epoch": 0.23990423374783182, + "grad_norm": 0.007241689134389162, + "learning_rate": 0.2889886497721103, + "loss": 0.1715, + "num_input_tokens_seen": 6368576, + "step": 4910 + }, + { + "epoch": 0.24014853541152614, + "grad_norm": 0.005694471765309572, + "learning_rate": 0.28896648666690505, + "loss": 0.1453, + "num_input_tokens_seen": 6374656, + "step": 4915 + }, + { + "epoch": 0.2403928370752205, + "grad_norm": 0.002581710461527109, + "learning_rate": 0.2889443021313212, + "loss": 0.1499, + "num_input_tokens_seen": 6381312, + "step": 4920 + }, + { + "epoch": 0.24063713873891482, + "grad_norm": 0.003219661768525839, + "learning_rate": 0.28892209616877984, + "loss": 0.1146, + "num_input_tokens_seen": 6388224, + "step": 4925 + }, + { + "epoch": 0.24088144040260914, + "grad_norm": 0.003194965422153473, + "learning_rate": 0.28889986878270546, + "loss": 0.1253, + "num_input_tokens_seen": 6394336, + "step": 4930 + }, + { + "epoch": 0.24112574206630347, + "grad_norm": 0.003228870453312993, + "learning_rate": 0.28887761997652583, + "loss": 0.1456, + "num_input_tokens_seen": 6400512, + "step": 4935 + }, + { + "epoch": 0.2413700437299978, + "grad_norm": 0.002667984925210476, + "learning_rate": 0.2888553497536719, + "loss": 0.1472, + "num_input_tokens_seen": 6407200, + "step": 4940 + }, + { + "epoch": 0.24161434539369214, + "grad_norm": 0.0023510472383350134, + "learning_rate": 0.2888330581175781, + "loss": 0.1141, + "num_input_tokens_seen": 6413984, + "step": 4945 + }, + { + "epoch": 0.24185864705738647, + "grad_norm": 0.004507054109126329, + "learning_rate": 0.28881074507168203, + "loss": 0.1296, + "num_input_tokens_seen": 6420704, + "step": 4950 + }, + { + "epoch": 0.2421029487210808, + "grad_norm": 0.0038737363647669554, + "learning_rate": 0.2887884106194247, + "loss": 0.1265, + "num_input_tokens_seen": 6427168, + "step": 4955 + }, + { + "epoch": 0.24234725038477511, + "grad_norm": 0.005534245166927576, + "learning_rate": 0.28876605476425027, + "loss": 0.1258, + "num_input_tokens_seen": 6433664, + "step": 4960 + }, + { + "epoch": 0.24259155204846944, + "grad_norm": 0.005809430964291096, + "learning_rate": 0.2887436775096064, + "loss": 0.1228, + "num_input_tokens_seen": 6439744, + "step": 4965 + }, + { + "epoch": 0.2428358537121638, + "grad_norm": 0.0035656532272696495, + "learning_rate": 0.2887212788589439, + "loss": 0.114, + "num_input_tokens_seen": 6446528, + "step": 4970 + }, + { + "epoch": 0.24308015537585811, + "grad_norm": 0.007235658820718527, + "learning_rate": 0.2886988588157169, + "loss": 0.1297, + "num_input_tokens_seen": 6452896, + "step": 4975 + }, + { + "epoch": 0.24332445703955244, + "grad_norm": 0.005651682615280151, + "learning_rate": 0.28867641738338284, + "loss": 0.1458, + "num_input_tokens_seen": 6459424, + "step": 4980 + }, + { + "epoch": 0.24356875870324676, + "grad_norm": 0.005846000276505947, + "learning_rate": 0.2886539545654026, + "loss": 0.1349, + "num_input_tokens_seen": 6466208, + "step": 4985 + }, + { + "epoch": 0.2438130603669411, + "grad_norm": 0.003531825728714466, + "learning_rate": 0.28863147036524006, + "loss": 0.1282, + "num_input_tokens_seen": 6473152, + "step": 4990 + }, + { + "epoch": 0.24405736203063544, + "grad_norm": 0.0032157679088413715, + "learning_rate": 0.2886089647863626, + "loss": 0.0906, + "num_input_tokens_seen": 6479680, + "step": 4995 + }, + { + "epoch": 0.24430166369432976, + "grad_norm": 0.003658934962004423, + "learning_rate": 0.288586437832241, + "loss": 0.111, + "num_input_tokens_seen": 6486336, + "step": 5000 + }, + { + "epoch": 0.24430166369432976, + "eval_loss": 0.13413235545158386, + "eval_runtime": 402.2105, + "eval_samples_per_second": 90.463, + "eval_steps_per_second": 22.618, + "num_input_tokens_seen": 6486336, + "step": 5000 + }, + { + "epoch": 0.2445459653580241, + "grad_norm": 0.005628323648124933, + "learning_rate": 0.28856388950634904, + "loss": 0.1356, + "num_input_tokens_seen": 6492992, + "step": 5005 + }, + { + "epoch": 0.2447902670217184, + "grad_norm": 0.004833749961107969, + "learning_rate": 0.288541319812164, + "loss": 0.1208, + "num_input_tokens_seen": 6499520, + "step": 5010 + }, + { + "epoch": 0.24503456868541273, + "grad_norm": 0.0063050673343241215, + "learning_rate": 0.2885187287531665, + "loss": 0.1181, + "num_input_tokens_seen": 6505568, + "step": 5015 + }, + { + "epoch": 0.2452788703491071, + "grad_norm": 0.0061167161911726, + "learning_rate": 0.2884961163328402, + "loss": 0.115, + "num_input_tokens_seen": 6512160, + "step": 5020 + }, + { + "epoch": 0.2455231720128014, + "grad_norm": 0.007771620526909828, + "learning_rate": 0.28847348255467237, + "loss": 0.1279, + "num_input_tokens_seen": 6518144, + "step": 5025 + }, + { + "epoch": 0.24576747367649573, + "grad_norm": 0.005182047374546528, + "learning_rate": 0.28845082742215333, + "loss": 0.1247, + "num_input_tokens_seen": 6524576, + "step": 5030 + }, + { + "epoch": 0.24601177534019006, + "grad_norm": 0.004479315131902695, + "learning_rate": 0.2884281509387769, + "loss": 0.1351, + "num_input_tokens_seen": 6531008, + "step": 5035 + }, + { + "epoch": 0.24625607700388438, + "grad_norm": 0.004232538864016533, + "learning_rate": 0.2884054531080399, + "loss": 0.1356, + "num_input_tokens_seen": 6537344, + "step": 5040 + }, + { + "epoch": 0.24650037866757873, + "grad_norm": 0.00310306902974844, + "learning_rate": 0.28838273393344277, + "loss": 0.1249, + "num_input_tokens_seen": 6543904, + "step": 5045 + }, + { + "epoch": 0.24674468033127306, + "grad_norm": 0.004674668423831463, + "learning_rate": 0.288359993418489, + "loss": 0.1184, + "num_input_tokens_seen": 6550272, + "step": 5050 + }, + { + "epoch": 0.24698898199496738, + "grad_norm": 0.005113111808896065, + "learning_rate": 0.28833723156668556, + "loss": 0.116, + "num_input_tokens_seen": 6556608, + "step": 5055 + }, + { + "epoch": 0.2472332836586617, + "grad_norm": 0.003800424747169018, + "learning_rate": 0.2883144483815425, + "loss": 0.1258, + "num_input_tokens_seen": 6563168, + "step": 5060 + }, + { + "epoch": 0.24747758532235606, + "grad_norm": 0.0032585677690804005, + "learning_rate": 0.28829164386657335, + "loss": 0.121, + "num_input_tokens_seen": 6569472, + "step": 5065 + }, + { + "epoch": 0.24772188698605038, + "grad_norm": 0.006799273192882538, + "learning_rate": 0.28826881802529486, + "loss": 0.142, + "num_input_tokens_seen": 6575968, + "step": 5070 + }, + { + "epoch": 0.2479661886497447, + "grad_norm": 0.004573218058794737, + "learning_rate": 0.28824597086122705, + "loss": 0.1228, + "num_input_tokens_seen": 6582112, + "step": 5075 + }, + { + "epoch": 0.24821049031343903, + "grad_norm": 0.0045999204739928246, + "learning_rate": 0.28822310237789317, + "loss": 0.1249, + "num_input_tokens_seen": 6588576, + "step": 5080 + }, + { + "epoch": 0.24845479197713335, + "grad_norm": 0.0065007517114281654, + "learning_rate": 0.2882002125788199, + "loss": 0.1349, + "num_input_tokens_seen": 6594816, + "step": 5085 + }, + { + "epoch": 0.2486990936408277, + "grad_norm": 0.005319612100720406, + "learning_rate": 0.2881773014675371, + "loss": 0.144, + "num_input_tokens_seen": 6601184, + "step": 5090 + }, + { + "epoch": 0.24894339530452203, + "grad_norm": 0.003215676872059703, + "learning_rate": 0.288154369047578, + "loss": 0.1314, + "num_input_tokens_seen": 6607616, + "step": 5095 + }, + { + "epoch": 0.24918769696821635, + "grad_norm": 0.0030264032538980246, + "learning_rate": 0.28813141532247905, + "loss": 0.1315, + "num_input_tokens_seen": 6613984, + "step": 5100 + }, + { + "epoch": 0.24943199863191068, + "grad_norm": 0.0036272285506129265, + "learning_rate": 0.28810844029578, + "loss": 0.1268, + "num_input_tokens_seen": 6620544, + "step": 5105 + }, + { + "epoch": 0.249676300295605, + "grad_norm": 0.004190557636320591, + "learning_rate": 0.2880854439710238, + "loss": 0.1473, + "num_input_tokens_seen": 6627232, + "step": 5110 + }, + { + "epoch": 0.24992060195929935, + "grad_norm": 0.0036716845352202654, + "learning_rate": 0.28806242635175694, + "loss": 0.1313, + "num_input_tokens_seen": 6633728, + "step": 5115 + }, + { + "epoch": 0.25016490362299365, + "grad_norm": 0.005418386310338974, + "learning_rate": 0.2880393874415289, + "loss": 0.1274, + "num_input_tokens_seen": 6640288, + "step": 5120 + }, + { + "epoch": 0.250409205286688, + "grad_norm": 0.0022259212564677, + "learning_rate": 0.2880163272438926, + "loss": 0.1234, + "num_input_tokens_seen": 6647072, + "step": 5125 + }, + { + "epoch": 0.25065350695038235, + "grad_norm": 0.0023107435554265976, + "learning_rate": 0.2879932457624042, + "loss": 0.116, + "num_input_tokens_seen": 6654112, + "step": 5130 + }, + { + "epoch": 0.2508978086140767, + "grad_norm": 0.006231924053281546, + "learning_rate": 0.2879701430006232, + "loss": 0.1573, + "num_input_tokens_seen": 6661056, + "step": 5135 + }, + { + "epoch": 0.251142110277771, + "grad_norm": 0.0034895702265203, + "learning_rate": 0.28794701896211233, + "loss": 0.1775, + "num_input_tokens_seen": 6667648, + "step": 5140 + }, + { + "epoch": 0.2513864119414653, + "grad_norm": 0.004335740115493536, + "learning_rate": 0.28792387365043753, + "loss": 0.1309, + "num_input_tokens_seen": 6674208, + "step": 5145 + }, + { + "epoch": 0.25163071360515965, + "grad_norm": 0.004260957706719637, + "learning_rate": 0.28790070706916815, + "loss": 0.1142, + "num_input_tokens_seen": 6680544, + "step": 5150 + }, + { + "epoch": 0.251875015268854, + "grad_norm": 0.003159480867907405, + "learning_rate": 0.2878775192218768, + "loss": 0.1426, + "num_input_tokens_seen": 6687008, + "step": 5155 + }, + { + "epoch": 0.2521193169325483, + "grad_norm": 0.00369929987937212, + "learning_rate": 0.2878543101121393, + "loss": 0.1487, + "num_input_tokens_seen": 6693248, + "step": 5160 + }, + { + "epoch": 0.2523636185962426, + "grad_norm": 0.003381680464372039, + "learning_rate": 0.28783107974353483, + "loss": 0.1322, + "num_input_tokens_seen": 6700064, + "step": 5165 + }, + { + "epoch": 0.25260792025993695, + "grad_norm": 0.00345210125669837, + "learning_rate": 0.2878078281196457, + "loss": 0.1237, + "num_input_tokens_seen": 6706528, + "step": 5170 + }, + { + "epoch": 0.2528522219236313, + "grad_norm": 0.00475513469427824, + "learning_rate": 0.28778455524405777, + "loss": 0.1047, + "num_input_tokens_seen": 6712832, + "step": 5175 + }, + { + "epoch": 0.25309652358732565, + "grad_norm": 0.009362848475575447, + "learning_rate": 0.2877612611203598, + "loss": 0.1115, + "num_input_tokens_seen": 6719200, + "step": 5180 + }, + { + "epoch": 0.25334082525102, + "grad_norm": 0.007876661606132984, + "learning_rate": 0.28773794575214423, + "loss": 0.1434, + "num_input_tokens_seen": 6725440, + "step": 5185 + }, + { + "epoch": 0.2535851269147143, + "grad_norm": 0.005508119240403175, + "learning_rate": 0.28771460914300645, + "loss": 0.1343, + "num_input_tokens_seen": 6732064, + "step": 5190 + }, + { + "epoch": 0.2538294285784086, + "grad_norm": 0.01184593141078949, + "learning_rate": 0.2876912512965454, + "loss": 0.1496, + "num_input_tokens_seen": 6738432, + "step": 5195 + }, + { + "epoch": 0.25407373024210295, + "grad_norm": 0.005316245835274458, + "learning_rate": 0.287667872216363, + "loss": 0.1272, + "num_input_tokens_seen": 6744864, + "step": 5200 + }, + { + "epoch": 0.25407373024210295, + "eval_loss": 0.13471271097660065, + "eval_runtime": 402.3136, + "eval_samples_per_second": 90.439, + "eval_steps_per_second": 22.612, + "num_input_tokens_seen": 6744864, + "step": 5200 + }, + { + "epoch": 0.25431803190579727, + "grad_norm": 0.004070569761097431, + "learning_rate": 0.2876444719060647, + "loss": 0.1356, + "num_input_tokens_seen": 6751328, + "step": 5205 + }, + { + "epoch": 0.2545623335694916, + "grad_norm": 0.007996279746294022, + "learning_rate": 0.287621050369259, + "loss": 0.1331, + "num_input_tokens_seen": 6757504, + "step": 5210 + }, + { + "epoch": 0.2548066352331859, + "grad_norm": 0.003846957581117749, + "learning_rate": 0.28759760760955794, + "loss": 0.1347, + "num_input_tokens_seen": 6763744, + "step": 5215 + }, + { + "epoch": 0.25505093689688024, + "grad_norm": 0.0035930154845118523, + "learning_rate": 0.2875741436305766, + "loss": 0.1099, + "num_input_tokens_seen": 6770400, + "step": 5220 + }, + { + "epoch": 0.2552952385605746, + "grad_norm": 0.005588141269981861, + "learning_rate": 0.28755065843593347, + "loss": 0.1325, + "num_input_tokens_seen": 6776960, + "step": 5225 + }, + { + "epoch": 0.25553954022426895, + "grad_norm": 0.009593975730240345, + "learning_rate": 0.2875271520292502, + "loss": 0.1321, + "num_input_tokens_seen": 6783328, + "step": 5230 + }, + { + "epoch": 0.25578384188796327, + "grad_norm": 0.008006543852388859, + "learning_rate": 0.28750362441415184, + "loss": 0.1342, + "num_input_tokens_seen": 6789440, + "step": 5235 + }, + { + "epoch": 0.2560281435516576, + "grad_norm": 0.004480399657040834, + "learning_rate": 0.28748007559426664, + "loss": 0.1241, + "num_input_tokens_seen": 6796832, + "step": 5240 + }, + { + "epoch": 0.2562724452153519, + "grad_norm": 0.004550971556454897, + "learning_rate": 0.2874565055732261, + "loss": 0.1168, + "num_input_tokens_seen": 6803328, + "step": 5245 + }, + { + "epoch": 0.25651674687904624, + "grad_norm": 0.010320568457245827, + "learning_rate": 0.28743291435466495, + "loss": 0.142, + "num_input_tokens_seen": 6809728, + "step": 5250 + }, + { + "epoch": 0.25676104854274057, + "grad_norm": 0.005282815545797348, + "learning_rate": 0.2874093019422214, + "loss": 0.1407, + "num_input_tokens_seen": 6816448, + "step": 5255 + }, + { + "epoch": 0.2570053502064349, + "grad_norm": 0.0038880875799804926, + "learning_rate": 0.28738566833953666, + "loss": 0.1336, + "num_input_tokens_seen": 6823040, + "step": 5260 + }, + { + "epoch": 0.2572496518701292, + "grad_norm": 0.004238972906023264, + "learning_rate": 0.28736201355025537, + "loss": 0.1464, + "num_input_tokens_seen": 6829600, + "step": 5265 + }, + { + "epoch": 0.25749395353382354, + "grad_norm": 0.0034397707786411047, + "learning_rate": 0.28733833757802535, + "loss": 0.1451, + "num_input_tokens_seen": 6836128, + "step": 5270 + }, + { + "epoch": 0.2577382551975179, + "grad_norm": 0.0021710984874516726, + "learning_rate": 0.28731464042649785, + "loss": 0.1081, + "num_input_tokens_seen": 6842944, + "step": 5275 + }, + { + "epoch": 0.25798255686121224, + "grad_norm": 0.002989856991916895, + "learning_rate": 0.2872909220993271, + "loss": 0.1143, + "num_input_tokens_seen": 6849792, + "step": 5280 + }, + { + "epoch": 0.25822685852490657, + "grad_norm": 0.0030976496636867523, + "learning_rate": 0.287267182600171, + "loss": 0.1317, + "num_input_tokens_seen": 6856064, + "step": 5285 + }, + { + "epoch": 0.2584711601886009, + "grad_norm": 0.003915499430149794, + "learning_rate": 0.2872434219326902, + "loss": 0.1301, + "num_input_tokens_seen": 6862272, + "step": 5290 + }, + { + "epoch": 0.2587154618522952, + "grad_norm": 0.00699248630553484, + "learning_rate": 0.28721964010054907, + "loss": 0.1546, + "num_input_tokens_seen": 6869120, + "step": 5295 + }, + { + "epoch": 0.25895976351598954, + "grad_norm": 0.007647676859050989, + "learning_rate": 0.28719583710741503, + "loss": 0.1489, + "num_input_tokens_seen": 6875712, + "step": 5300 + }, + { + "epoch": 0.25920406517968386, + "grad_norm": 0.004205972421914339, + "learning_rate": 0.28717201295695877, + "loss": 0.1303, + "num_input_tokens_seen": 6882784, + "step": 5305 + }, + { + "epoch": 0.2594483668433782, + "grad_norm": 0.006530147045850754, + "learning_rate": 0.28714816765285434, + "loss": 0.1479, + "num_input_tokens_seen": 6889056, + "step": 5310 + }, + { + "epoch": 0.2596926685070725, + "grad_norm": 0.006116996519267559, + "learning_rate": 0.28712430119877896, + "loss": 0.1455, + "num_input_tokens_seen": 6895392, + "step": 5315 + }, + { + "epoch": 0.2599369701707669, + "grad_norm": 0.004145187325775623, + "learning_rate": 0.28710041359841304, + "loss": 0.1429, + "num_input_tokens_seen": 6901760, + "step": 5320 + }, + { + "epoch": 0.2601812718344612, + "grad_norm": 0.005178912542760372, + "learning_rate": 0.28707650485544056, + "loss": 0.1303, + "num_input_tokens_seen": 6908800, + "step": 5325 + }, + { + "epoch": 0.26042557349815554, + "grad_norm": 0.0027162127662450075, + "learning_rate": 0.28705257497354836, + "loss": 0.1212, + "num_input_tokens_seen": 6915040, + "step": 5330 + }, + { + "epoch": 0.26066987516184986, + "grad_norm": 0.005336425267159939, + "learning_rate": 0.28702862395642675, + "loss": 0.1289, + "num_input_tokens_seen": 6921344, + "step": 5335 + }, + { + "epoch": 0.2609141768255442, + "grad_norm": 0.004297249484807253, + "learning_rate": 0.28700465180776935, + "loss": 0.1413, + "num_input_tokens_seen": 6928480, + "step": 5340 + }, + { + "epoch": 0.2611584784892385, + "grad_norm": 0.004572005942463875, + "learning_rate": 0.2869806585312729, + "loss": 0.1394, + "num_input_tokens_seen": 6934976, + "step": 5345 + }, + { + "epoch": 0.26140278015293283, + "grad_norm": 0.0054540145210921764, + "learning_rate": 0.28695664413063754, + "loss": 0.1373, + "num_input_tokens_seen": 6941600, + "step": 5350 + }, + { + "epoch": 0.26164708181662716, + "grad_norm": 0.003927123732864857, + "learning_rate": 0.28693260860956654, + "loss": 0.1693, + "num_input_tokens_seen": 6947904, + "step": 5355 + }, + { + "epoch": 0.2618913834803215, + "grad_norm": 0.003003922523930669, + "learning_rate": 0.2869085519717665, + "loss": 0.1316, + "num_input_tokens_seen": 6954624, + "step": 5360 + }, + { + "epoch": 0.2621356851440158, + "grad_norm": 0.00245377398096025, + "learning_rate": 0.28688447422094726, + "loss": 0.1264, + "num_input_tokens_seen": 6961152, + "step": 5365 + }, + { + "epoch": 0.2623799868077102, + "grad_norm": 0.003575288923457265, + "learning_rate": 0.2868603753608219, + "loss": 0.1386, + "num_input_tokens_seen": 6967584, + "step": 5370 + }, + { + "epoch": 0.2626242884714045, + "grad_norm": 0.0027833881322294474, + "learning_rate": 0.28683625539510665, + "loss": 0.1359, + "num_input_tokens_seen": 6973952, + "step": 5375 + }, + { + "epoch": 0.26286859013509883, + "grad_norm": 0.002473176224157214, + "learning_rate": 0.28681211432752135, + "loss": 0.1308, + "num_input_tokens_seen": 6980352, + "step": 5380 + }, + { + "epoch": 0.26311289179879316, + "grad_norm": 0.003131711622700095, + "learning_rate": 0.2867879521617887, + "loss": 0.134, + "num_input_tokens_seen": 6986752, + "step": 5385 + }, + { + "epoch": 0.2633571934624875, + "grad_norm": 0.003886363236233592, + "learning_rate": 0.28676376890163485, + "loss": 0.1202, + "num_input_tokens_seen": 6993280, + "step": 5390 + }, + { + "epoch": 0.2636014951261818, + "grad_norm": 0.0033159193117171526, + "learning_rate": 0.2867395645507891, + "loss": 0.122, + "num_input_tokens_seen": 6999904, + "step": 5395 + }, + { + "epoch": 0.26384579678987613, + "grad_norm": 0.006251761224120855, + "learning_rate": 0.2867153391129842, + "loss": 0.1251, + "num_input_tokens_seen": 7006944, + "step": 5400 + }, + { + "epoch": 0.26384579678987613, + "eval_loss": 0.1302129626274109, + "eval_runtime": 402.3891, + "eval_samples_per_second": 90.422, + "eval_steps_per_second": 22.607, + "num_input_tokens_seen": 7006944, + "step": 5400 + }, + { + "epoch": 0.26409009845357045, + "grad_norm": 0.005955981090664864, + "learning_rate": 0.28669109259195585, + "loss": 0.1247, + "num_input_tokens_seen": 7013632, + "step": 5405 + }, + { + "epoch": 0.2643344001172648, + "grad_norm": 0.004658418241888285, + "learning_rate": 0.2866668249914433, + "loss": 0.1396, + "num_input_tokens_seen": 7020192, + "step": 5410 + }, + { + "epoch": 0.2645787017809591, + "grad_norm": 0.004894699435681105, + "learning_rate": 0.2866425363151889, + "loss": 0.1258, + "num_input_tokens_seen": 7026400, + "step": 5415 + }, + { + "epoch": 0.2648230034446535, + "grad_norm": 0.003387420903891325, + "learning_rate": 0.2866182265669382, + "loss": 0.1074, + "num_input_tokens_seen": 7032544, + "step": 5420 + }, + { + "epoch": 0.2650673051083478, + "grad_norm": 0.00541591364890337, + "learning_rate": 0.28659389575044014, + "loss": 0.1346, + "num_input_tokens_seen": 7039136, + "step": 5425 + }, + { + "epoch": 0.26531160677204213, + "grad_norm": 0.0065269251354038715, + "learning_rate": 0.28656954386944683, + "loss": 0.1278, + "num_input_tokens_seen": 7045152, + "step": 5430 + }, + { + "epoch": 0.26555590843573645, + "grad_norm": 0.00591407110914588, + "learning_rate": 0.28654517092771353, + "loss": 0.1217, + "num_input_tokens_seen": 7051680, + "step": 5435 + }, + { + "epoch": 0.2658002100994308, + "grad_norm": 0.004400928970426321, + "learning_rate": 0.286520776928999, + "loss": 0.1147, + "num_input_tokens_seen": 7058112, + "step": 5440 + }, + { + "epoch": 0.2660445117631251, + "grad_norm": 0.004510826896876097, + "learning_rate": 0.286496361877065, + "loss": 0.1352, + "num_input_tokens_seen": 7064832, + "step": 5445 + }, + { + "epoch": 0.2662888134268194, + "grad_norm": 0.0031608904246240854, + "learning_rate": 0.28647192577567676, + "loss": 0.1363, + "num_input_tokens_seen": 7071520, + "step": 5450 + }, + { + "epoch": 0.26653311509051375, + "grad_norm": 0.006053500808775425, + "learning_rate": 0.28644746862860254, + "loss": 0.1296, + "num_input_tokens_seen": 7078016, + "step": 5455 + }, + { + "epoch": 0.2667774167542081, + "grad_norm": 0.0035545264836400747, + "learning_rate": 0.2864229904396139, + "loss": 0.1276, + "num_input_tokens_seen": 7084576, + "step": 5460 + }, + { + "epoch": 0.2670217184179024, + "grad_norm": 0.004954153206199408, + "learning_rate": 0.28639849121248573, + "loss": 0.1434, + "num_input_tokens_seen": 7091360, + "step": 5465 + }, + { + "epoch": 0.2672660200815968, + "grad_norm": 0.003700742730870843, + "learning_rate": 0.28637397095099615, + "loss": 0.1356, + "num_input_tokens_seen": 7098016, + "step": 5470 + }, + { + "epoch": 0.2675103217452911, + "grad_norm": 0.0026276977732777596, + "learning_rate": 0.28634942965892646, + "loss": 0.1316, + "num_input_tokens_seen": 7104704, + "step": 5475 + }, + { + "epoch": 0.2677546234089854, + "grad_norm": 0.003253340721130371, + "learning_rate": 0.28632486734006124, + "loss": 0.1078, + "num_input_tokens_seen": 7111360, + "step": 5480 + }, + { + "epoch": 0.26799892507267975, + "grad_norm": 0.0072964574210345745, + "learning_rate": 0.28630028399818835, + "loss": 0.1396, + "num_input_tokens_seen": 7117728, + "step": 5485 + }, + { + "epoch": 0.2682432267363741, + "grad_norm": 0.0034082653000950813, + "learning_rate": 0.2862756796370987, + "loss": 0.0928, + "num_input_tokens_seen": 7124192, + "step": 5490 + }, + { + "epoch": 0.2684875284000684, + "grad_norm": 0.005915261805057526, + "learning_rate": 0.2862510542605868, + "loss": 0.1429, + "num_input_tokens_seen": 7131072, + "step": 5495 + }, + { + "epoch": 0.2687318300637627, + "grad_norm": 0.004800623282790184, + "learning_rate": 0.2862264078724501, + "loss": 0.1367, + "num_input_tokens_seen": 7137504, + "step": 5500 + }, + { + "epoch": 0.26897613172745705, + "grad_norm": 0.0043573444709181786, + "learning_rate": 0.28620174047648933, + "loss": 0.1441, + "num_input_tokens_seen": 7143904, + "step": 5505 + }, + { + "epoch": 0.26922043339115137, + "grad_norm": 0.006739671342074871, + "learning_rate": 0.2861770520765086, + "loss": 0.1225, + "num_input_tokens_seen": 7150208, + "step": 5510 + }, + { + "epoch": 0.26946473505484575, + "grad_norm": 0.007493209093809128, + "learning_rate": 0.2861523426763151, + "loss": 0.126, + "num_input_tokens_seen": 7157088, + "step": 5515 + }, + { + "epoch": 0.2697090367185401, + "grad_norm": 0.005771221127361059, + "learning_rate": 0.2861276122797194, + "loss": 0.13, + "num_input_tokens_seen": 7163904, + "step": 5520 + }, + { + "epoch": 0.2699533383822344, + "grad_norm": 0.004787174519151449, + "learning_rate": 0.28610286089053516, + "loss": 0.1432, + "num_input_tokens_seen": 7170176, + "step": 5525 + }, + { + "epoch": 0.2701976400459287, + "grad_norm": 0.0031659642700105906, + "learning_rate": 0.28607808851257943, + "loss": 0.1503, + "num_input_tokens_seen": 7176864, + "step": 5530 + }, + { + "epoch": 0.27044194170962305, + "grad_norm": 0.005717766936868429, + "learning_rate": 0.28605329514967237, + "loss": 0.126, + "num_input_tokens_seen": 7183520, + "step": 5535 + }, + { + "epoch": 0.27068624337331737, + "grad_norm": 0.0038151063490659, + "learning_rate": 0.2860284808056374, + "loss": 0.1312, + "num_input_tokens_seen": 7190112, + "step": 5540 + }, + { + "epoch": 0.2709305450370117, + "grad_norm": 0.002949526533484459, + "learning_rate": 0.28600364548430135, + "loss": 0.1074, + "num_input_tokens_seen": 7196448, + "step": 5545 + }, + { + "epoch": 0.271174846700706, + "grad_norm": 0.0024889507330954075, + "learning_rate": 0.28597878918949393, + "loss": 0.1172, + "num_input_tokens_seen": 7202816, + "step": 5550 + }, + { + "epoch": 0.27141914836440034, + "grad_norm": 0.0028458135202527046, + "learning_rate": 0.2859539119250485, + "loss": 0.1128, + "num_input_tokens_seen": 7209408, + "step": 5555 + }, + { + "epoch": 0.27166345002809467, + "grad_norm": 0.007140621542930603, + "learning_rate": 0.2859290136948013, + "loss": 0.1092, + "num_input_tokens_seen": 7215904, + "step": 5560 + }, + { + "epoch": 0.27190775169178905, + "grad_norm": 0.00449804961681366, + "learning_rate": 0.28590409450259197, + "loss": 0.1239, + "num_input_tokens_seen": 7222240, + "step": 5565 + }, + { + "epoch": 0.27215205335548337, + "grad_norm": 0.00206343294121325, + "learning_rate": 0.28587915435226346, + "loss": 0.0903, + "num_input_tokens_seen": 7228608, + "step": 5570 + }, + { + "epoch": 0.2723963550191777, + "grad_norm": 0.008042197674512863, + "learning_rate": 0.2858541932476617, + "loss": 0.1147, + "num_input_tokens_seen": 7235744, + "step": 5575 + }, + { + "epoch": 0.272640656682872, + "grad_norm": 0.0058980765752494335, + "learning_rate": 0.2858292111926361, + "loss": 0.1345, + "num_input_tokens_seen": 7242176, + "step": 5580 + }, + { + "epoch": 0.27288495834656634, + "grad_norm": 0.004693014547228813, + "learning_rate": 0.28580420819103924, + "loss": 0.1352, + "num_input_tokens_seen": 7248192, + "step": 5585 + }, + { + "epoch": 0.27312926001026067, + "grad_norm": 0.0027627339586615562, + "learning_rate": 0.2857791842467269, + "loss": 0.1194, + "num_input_tokens_seen": 7254720, + "step": 5590 + }, + { + "epoch": 0.273373561673955, + "grad_norm": 0.004043642897158861, + "learning_rate": 0.2857541393635579, + "loss": 0.1336, + "num_input_tokens_seen": 7261088, + "step": 5595 + }, + { + "epoch": 0.2736178633376493, + "grad_norm": 0.0037648712750524282, + "learning_rate": 0.2857290735453948, + "loss": 0.125, + "num_input_tokens_seen": 7267584, + "step": 5600 + }, + { + "epoch": 0.2736178633376493, + "eval_loss": 0.12559133768081665, + "eval_runtime": 402.7295, + "eval_samples_per_second": 90.346, + "eval_steps_per_second": 22.588, + "num_input_tokens_seen": 7267584, + "step": 5600 + }, + { + "epoch": 0.27386216500134364, + "grad_norm": 0.0053334664553403854, + "learning_rate": 0.28570398679610276, + "loss": 0.1067, + "num_input_tokens_seen": 7274080, + "step": 5605 + }, + { + "epoch": 0.27410646666503796, + "grad_norm": 0.005375921726226807, + "learning_rate": 0.2856788791195506, + "loss": 0.1437, + "num_input_tokens_seen": 7281824, + "step": 5610 + }, + { + "epoch": 0.27435076832873234, + "grad_norm": 0.007899930700659752, + "learning_rate": 0.28565375051961023, + "loss": 0.1297, + "num_input_tokens_seen": 7288384, + "step": 5615 + }, + { + "epoch": 0.27459506999242667, + "grad_norm": 0.003922676667571068, + "learning_rate": 0.28562860100015686, + "loss": 0.1161, + "num_input_tokens_seen": 7294976, + "step": 5620 + }, + { + "epoch": 0.274839371656121, + "grad_norm": 0.046299662441015244, + "learning_rate": 0.2856034305650687, + "loss": 0.1396, + "num_input_tokens_seen": 7301568, + "step": 5625 + }, + { + "epoch": 0.2750836733198153, + "grad_norm": 0.01735769398510456, + "learning_rate": 0.28557823921822756, + "loss": 0.1806, + "num_input_tokens_seen": 7308192, + "step": 5630 + }, + { + "epoch": 0.27532797498350964, + "grad_norm": 0.019972674548625946, + "learning_rate": 0.2855530269635181, + "loss": 0.2099, + "num_input_tokens_seen": 7314432, + "step": 5635 + }, + { + "epoch": 0.27557227664720396, + "grad_norm": 0.005862708203494549, + "learning_rate": 0.2855277938048284, + "loss": 0.1799, + "num_input_tokens_seen": 7320864, + "step": 5640 + }, + { + "epoch": 0.2758165783108983, + "grad_norm": 0.011841526255011559, + "learning_rate": 0.2855025397460498, + "loss": 0.1645, + "num_input_tokens_seen": 7327904, + "step": 5645 + }, + { + "epoch": 0.2760608799745926, + "grad_norm": 0.024735702201724052, + "learning_rate": 0.28547726479107666, + "loss": 0.1925, + "num_input_tokens_seen": 7334208, + "step": 5650 + }, + { + "epoch": 0.27630518163828693, + "grad_norm": 0.0026390203274786472, + "learning_rate": 0.2854519689438068, + "loss": 0.1736, + "num_input_tokens_seen": 7340736, + "step": 5655 + }, + { + "epoch": 0.2765494833019813, + "grad_norm": 0.0053466251119971275, + "learning_rate": 0.2854266522081412, + "loss": 0.1589, + "num_input_tokens_seen": 7347488, + "step": 5660 + }, + { + "epoch": 0.27679378496567564, + "grad_norm": 0.005495019257068634, + "learning_rate": 0.28540131458798385, + "loss": 0.1636, + "num_input_tokens_seen": 7353856, + "step": 5665 + }, + { + "epoch": 0.27703808662936996, + "grad_norm": 0.0036699676420539618, + "learning_rate": 0.28537595608724226, + "loss": 0.1757, + "num_input_tokens_seen": 7360192, + "step": 5670 + }, + { + "epoch": 0.2772823882930643, + "grad_norm": 0.0033017483074218035, + "learning_rate": 0.28535057670982705, + "loss": 0.1581, + "num_input_tokens_seen": 7366528, + "step": 5675 + }, + { + "epoch": 0.2775266899567586, + "grad_norm": 0.004743311088532209, + "learning_rate": 0.285325176459652, + "loss": 0.1842, + "num_input_tokens_seen": 7372672, + "step": 5680 + }, + { + "epoch": 0.27777099162045293, + "grad_norm": 0.002193101681768894, + "learning_rate": 0.28529975534063406, + "loss": 0.1832, + "num_input_tokens_seen": 7379456, + "step": 5685 + }, + { + "epoch": 0.27801529328414726, + "grad_norm": 0.0037478345911949873, + "learning_rate": 0.2852743133566936, + "loss": 0.1555, + "num_input_tokens_seen": 7385600, + "step": 5690 + }, + { + "epoch": 0.2782595949478416, + "grad_norm": 0.0027996350545436144, + "learning_rate": 0.2852488505117541, + "loss": 0.1675, + "num_input_tokens_seen": 7392224, + "step": 5695 + }, + { + "epoch": 0.2785038966115359, + "grad_norm": 0.002629923867061734, + "learning_rate": 0.28522336680974214, + "loss": 0.1625, + "num_input_tokens_seen": 7398656, + "step": 5700 + }, + { + "epoch": 0.27874819827523023, + "grad_norm": 0.0039253076538443565, + "learning_rate": 0.2851978622545877, + "loss": 0.1388, + "num_input_tokens_seen": 7405216, + "step": 5705 + }, + { + "epoch": 0.2789924999389246, + "grad_norm": 0.004253987688571215, + "learning_rate": 0.285172336850224, + "loss": 0.1177, + "num_input_tokens_seen": 7411584, + "step": 5710 + }, + { + "epoch": 0.27923680160261893, + "grad_norm": 0.004213154315948486, + "learning_rate": 0.2851467906005871, + "loss": 0.1557, + "num_input_tokens_seen": 7418016, + "step": 5715 + }, + { + "epoch": 0.27948110326631326, + "grad_norm": 0.005310391541570425, + "learning_rate": 0.28512122350961683, + "loss": 0.1561, + "num_input_tokens_seen": 7424384, + "step": 5720 + }, + { + "epoch": 0.2797254049300076, + "grad_norm": 0.005298037081956863, + "learning_rate": 0.2850956355812559, + "loss": 0.1327, + "num_input_tokens_seen": 7430912, + "step": 5725 + }, + { + "epoch": 0.2799697065937019, + "grad_norm": 0.005093451123684645, + "learning_rate": 0.28507002681945015, + "loss": 0.1781, + "num_input_tokens_seen": 7437504, + "step": 5730 + }, + { + "epoch": 0.28021400825739623, + "grad_norm": 0.005116675049066544, + "learning_rate": 0.28504439722814895, + "loss": 0.1673, + "num_input_tokens_seen": 7444480, + "step": 5735 + }, + { + "epoch": 0.28045830992109055, + "grad_norm": 0.0016365890624001622, + "learning_rate": 0.28501874681130457, + "loss": 0.1407, + "num_input_tokens_seen": 7450464, + "step": 5740 + }, + { + "epoch": 0.2807026115847849, + "grad_norm": 0.002895343815907836, + "learning_rate": 0.2849930755728727, + "loss": 0.1279, + "num_input_tokens_seen": 7457280, + "step": 5745 + }, + { + "epoch": 0.2809469132484792, + "grad_norm": 0.003031200962141156, + "learning_rate": 0.28496738351681217, + "loss": 0.1391, + "num_input_tokens_seen": 7463808, + "step": 5750 + }, + { + "epoch": 0.2811912149121735, + "grad_norm": 0.0046304636634886265, + "learning_rate": 0.284941670647085, + "loss": 0.1522, + "num_input_tokens_seen": 7469952, + "step": 5755 + }, + { + "epoch": 0.2814355165758679, + "grad_norm": 0.0031242750119417906, + "learning_rate": 0.2849159369676563, + "loss": 0.1503, + "num_input_tokens_seen": 7477088, + "step": 5760 + }, + { + "epoch": 0.28167981823956223, + "grad_norm": 0.0021155178546905518, + "learning_rate": 0.2848901824824948, + "loss": 0.155, + "num_input_tokens_seen": 7483392, + "step": 5765 + }, + { + "epoch": 0.28192411990325655, + "grad_norm": 0.0022575119510293007, + "learning_rate": 0.284864407195572, + "loss": 0.1675, + "num_input_tokens_seen": 7489568, + "step": 5770 + }, + { + "epoch": 0.2821684215669509, + "grad_norm": 0.0025158661883324385, + "learning_rate": 0.28483861111086284, + "loss": 0.1539, + "num_input_tokens_seen": 7496512, + "step": 5775 + }, + { + "epoch": 0.2824127232306452, + "grad_norm": 0.003578941337764263, + "learning_rate": 0.2848127942323453, + "loss": 0.1824, + "num_input_tokens_seen": 7503584, + "step": 5780 + }, + { + "epoch": 0.2826570248943395, + "grad_norm": 0.002583493944257498, + "learning_rate": 0.2847869565640007, + "loss": 0.1464, + "num_input_tokens_seen": 7510304, + "step": 5785 + }, + { + "epoch": 0.28290132655803385, + "grad_norm": 0.0017960165860131383, + "learning_rate": 0.2847610981098136, + "loss": 0.1483, + "num_input_tokens_seen": 7516576, + "step": 5790 + }, + { + "epoch": 0.2831456282217282, + "grad_norm": 0.002295809332281351, + "learning_rate": 0.2847352188737716, + "loss": 0.1676, + "num_input_tokens_seen": 7523104, + "step": 5795 + }, + { + "epoch": 0.2833899298854225, + "grad_norm": 0.0022400098387151957, + "learning_rate": 0.2847093188598658, + "loss": 0.1792, + "num_input_tokens_seen": 7529280, + "step": 5800 + }, + { + "epoch": 0.2833899298854225, + "eval_loss": 0.14302465319633484, + "eval_runtime": 402.2769, + "eval_samples_per_second": 90.448, + "eval_steps_per_second": 22.614, + "num_input_tokens_seen": 7529280, + "step": 5800 + }, + { + "epoch": 0.2836342315491168, + "grad_norm": 0.00530401524156332, + "learning_rate": 0.28468339807209003, + "loss": 0.1544, + "num_input_tokens_seen": 7536064, + "step": 5805 + }, + { + "epoch": 0.2838785332128112, + "grad_norm": 0.002451085252687335, + "learning_rate": 0.2846574565144418, + "loss": 0.1542, + "num_input_tokens_seen": 7542688, + "step": 5810 + }, + { + "epoch": 0.2841228348765055, + "grad_norm": 0.0039650131948292255, + "learning_rate": 0.28463149419092154, + "loss": 0.1647, + "num_input_tokens_seen": 7548864, + "step": 5815 + }, + { + "epoch": 0.28436713654019985, + "grad_norm": 0.0029378836043179035, + "learning_rate": 0.284605511105533, + "loss": 0.1417, + "num_input_tokens_seen": 7555360, + "step": 5820 + }, + { + "epoch": 0.2846114382038942, + "grad_norm": 0.0036574790719896555, + "learning_rate": 0.28457950726228315, + "loss": 0.1185, + "num_input_tokens_seen": 7561952, + "step": 5825 + }, + { + "epoch": 0.2848557398675885, + "grad_norm": 0.002634722040966153, + "learning_rate": 0.28455348266518193, + "loss": 0.1472, + "num_input_tokens_seen": 7568416, + "step": 5830 + }, + { + "epoch": 0.2851000415312828, + "grad_norm": 0.003539833938702941, + "learning_rate": 0.28452743731824287, + "loss": 0.1355, + "num_input_tokens_seen": 7575008, + "step": 5835 + }, + { + "epoch": 0.28534434319497715, + "grad_norm": 0.0030225946102291346, + "learning_rate": 0.28450137122548236, + "loss": 0.1334, + "num_input_tokens_seen": 7581760, + "step": 5840 + }, + { + "epoch": 0.28558864485867147, + "grad_norm": 0.0022102987859398127, + "learning_rate": 0.2844752843909201, + "loss": 0.121, + "num_input_tokens_seen": 7588192, + "step": 5845 + }, + { + "epoch": 0.2858329465223658, + "grad_norm": 0.0038610845804214478, + "learning_rate": 0.28444917681857923, + "loss": 0.1506, + "num_input_tokens_seen": 7594304, + "step": 5850 + }, + { + "epoch": 0.2860772481860602, + "grad_norm": 0.0037222837563604116, + "learning_rate": 0.28442304851248557, + "loss": 0.1332, + "num_input_tokens_seen": 7600352, + "step": 5855 + }, + { + "epoch": 0.2863215498497545, + "grad_norm": 0.003949276637285948, + "learning_rate": 0.2843968994766686, + "loss": 0.1375, + "num_input_tokens_seen": 7607296, + "step": 5860 + }, + { + "epoch": 0.2865658515134488, + "grad_norm": 0.003001399338245392, + "learning_rate": 0.28437072971516075, + "loss": 0.1596, + "num_input_tokens_seen": 7614016, + "step": 5865 + }, + { + "epoch": 0.28681015317714315, + "grad_norm": 0.006335533689707518, + "learning_rate": 0.2843445392319979, + "loss": 0.156, + "num_input_tokens_seen": 7620320, + "step": 5870 + }, + { + "epoch": 0.28705445484083747, + "grad_norm": 0.0031090276315808296, + "learning_rate": 0.28431832803121865, + "loss": 0.1312, + "num_input_tokens_seen": 7627040, + "step": 5875 + }, + { + "epoch": 0.2872987565045318, + "grad_norm": 0.002834859536960721, + "learning_rate": 0.28429209611686534, + "loss": 0.1416, + "num_input_tokens_seen": 7633664, + "step": 5880 + }, + { + "epoch": 0.2875430581682261, + "grad_norm": 0.0045402017422020435, + "learning_rate": 0.28426584349298323, + "loss": 0.123, + "num_input_tokens_seen": 7640032, + "step": 5885 + }, + { + "epoch": 0.28778735983192044, + "grad_norm": 0.005272164475172758, + "learning_rate": 0.2842395701636207, + "loss": 0.1408, + "num_input_tokens_seen": 7646688, + "step": 5890 + }, + { + "epoch": 0.28803166149561477, + "grad_norm": 0.007197830826044083, + "learning_rate": 0.28421327613282954, + "loss": 0.1514, + "num_input_tokens_seen": 7653408, + "step": 5895 + }, + { + "epoch": 0.2882759631593091, + "grad_norm": 0.0030531056690961123, + "learning_rate": 0.28418696140466454, + "loss": 0.1607, + "num_input_tokens_seen": 7659648, + "step": 5900 + }, + { + "epoch": 0.28852026482300347, + "grad_norm": 0.004124704282730818, + "learning_rate": 0.2841606259831838, + "loss": 0.16, + "num_input_tokens_seen": 7666400, + "step": 5905 + }, + { + "epoch": 0.2887645664866978, + "grad_norm": 0.0026871180161833763, + "learning_rate": 0.2841342698724486, + "loss": 0.1558, + "num_input_tokens_seen": 7672768, + "step": 5910 + }, + { + "epoch": 0.2890088681503921, + "grad_norm": 0.006417717784643173, + "learning_rate": 0.28410789307652334, + "loss": 0.1641, + "num_input_tokens_seen": 7679680, + "step": 5915 + }, + { + "epoch": 0.28925316981408644, + "grad_norm": 0.0028656520880758762, + "learning_rate": 0.2840814955994756, + "loss": 0.164, + "num_input_tokens_seen": 7686240, + "step": 5920 + }, + { + "epoch": 0.28949747147778077, + "grad_norm": 0.0019340679282322526, + "learning_rate": 0.2840550774453763, + "loss": 0.1527, + "num_input_tokens_seen": 7692640, + "step": 5925 + }, + { + "epoch": 0.2897417731414751, + "grad_norm": 0.001999280881136656, + "learning_rate": 0.28402863861829947, + "loss": 0.1723, + "num_input_tokens_seen": 7699072, + "step": 5930 + }, + { + "epoch": 0.2899860748051694, + "grad_norm": 0.0022148003336042166, + "learning_rate": 0.2840021791223222, + "loss": 0.1452, + "num_input_tokens_seen": 7705248, + "step": 5935 + }, + { + "epoch": 0.29023037646886374, + "grad_norm": 0.003384433453902602, + "learning_rate": 0.2839756989615249, + "loss": 0.1584, + "num_input_tokens_seen": 7711968, + "step": 5940 + }, + { + "epoch": 0.29047467813255806, + "grad_norm": 0.0021649221889674664, + "learning_rate": 0.28394919813999125, + "loss": 0.1397, + "num_input_tokens_seen": 7718496, + "step": 5945 + }, + { + "epoch": 0.2907189797962524, + "grad_norm": 0.0024802943225950003, + "learning_rate": 0.28392267666180787, + "loss": 0.1342, + "num_input_tokens_seen": 7725440, + "step": 5950 + }, + { + "epoch": 0.29096328145994677, + "grad_norm": 0.0027223669458180666, + "learning_rate": 0.2838961345310648, + "loss": 0.1381, + "num_input_tokens_seen": 7731424, + "step": 5955 + }, + { + "epoch": 0.2912075831236411, + "grad_norm": 0.003419796470552683, + "learning_rate": 0.2838695717518552, + "loss": 0.1407, + "num_input_tokens_seen": 7737984, + "step": 5960 + }, + { + "epoch": 0.2914518847873354, + "grad_norm": 0.003257952630519867, + "learning_rate": 0.28384298832827526, + "loss": 0.1463, + "num_input_tokens_seen": 7744288, + "step": 5965 + }, + { + "epoch": 0.29169618645102974, + "grad_norm": 0.00265782093629241, + "learning_rate": 0.28381638426442457, + "loss": 0.1379, + "num_input_tokens_seen": 7750944, + "step": 5970 + }, + { + "epoch": 0.29194048811472406, + "grad_norm": 0.0023359558545053005, + "learning_rate": 0.2837897595644057, + "loss": 0.1511, + "num_input_tokens_seen": 7757024, + "step": 5975 + }, + { + "epoch": 0.2921847897784184, + "grad_norm": 0.006324345245957375, + "learning_rate": 0.28376311423232475, + "loss": 0.1262, + "num_input_tokens_seen": 7763360, + "step": 5980 + }, + { + "epoch": 0.2924290914421127, + "grad_norm": 0.004712031688541174, + "learning_rate": 0.2837364482722905, + "loss": 0.1528, + "num_input_tokens_seen": 7770144, + "step": 5985 + }, + { + "epoch": 0.29267339310580703, + "grad_norm": 0.0028683992568403482, + "learning_rate": 0.28370976168841533, + "loss": 0.152, + "num_input_tokens_seen": 7776192, + "step": 5990 + }, + { + "epoch": 0.29291769476950136, + "grad_norm": 0.001993064535781741, + "learning_rate": 0.2836830544848146, + "loss": 0.1339, + "num_input_tokens_seen": 7782368, + "step": 5995 + }, + { + "epoch": 0.29316199643319574, + "grad_norm": 0.0026106189470738173, + "learning_rate": 0.2836563266656069, + "loss": 0.1262, + "num_input_tokens_seen": 7788992, + "step": 6000 + }, + { + "epoch": 0.29316199643319574, + "eval_loss": 0.1473085582256317, + "eval_runtime": 402.1767, + "eval_samples_per_second": 90.47, + "eval_steps_per_second": 22.619, + "num_input_tokens_seen": 7788992, + "step": 6000 + }, + { + "epoch": 0.29340629809689006, + "grad_norm": 0.0028795951511710882, + "learning_rate": 0.283629578234914, + "loss": 0.1245, + "num_input_tokens_seen": 7795872, + "step": 6005 + }, + { + "epoch": 0.2936505997605844, + "grad_norm": 0.00349132320843637, + "learning_rate": 0.2836028091968608, + "loss": 0.1344, + "num_input_tokens_seen": 7802592, + "step": 6010 + }, + { + "epoch": 0.2938949014242787, + "grad_norm": 0.010197360068559647, + "learning_rate": 0.28357601955557554, + "loss": 0.1368, + "num_input_tokens_seen": 7809376, + "step": 6015 + }, + { + "epoch": 0.29413920308797303, + "grad_norm": 0.0029040079098194838, + "learning_rate": 0.2835492093151894, + "loss": 0.1612, + "num_input_tokens_seen": 7815872, + "step": 6020 + }, + { + "epoch": 0.29438350475166736, + "grad_norm": 0.0048814513720571995, + "learning_rate": 0.2835223784798369, + "loss": 0.1388, + "num_input_tokens_seen": 7822592, + "step": 6025 + }, + { + "epoch": 0.2946278064153617, + "grad_norm": 0.00239000516012311, + "learning_rate": 0.2834955270536557, + "loss": 0.1142, + "num_input_tokens_seen": 7829184, + "step": 6030 + }, + { + "epoch": 0.294872108079056, + "grad_norm": 0.004879494663327932, + "learning_rate": 0.2834686550407866, + "loss": 0.1192, + "num_input_tokens_seen": 7835744, + "step": 6035 + }, + { + "epoch": 0.29511640974275033, + "grad_norm": 0.0032097534276545048, + "learning_rate": 0.28344176244537367, + "loss": 0.1187, + "num_input_tokens_seen": 7842624, + "step": 6040 + }, + { + "epoch": 0.29536071140644465, + "grad_norm": 0.004066155292093754, + "learning_rate": 0.28341484927156396, + "loss": 0.1135, + "num_input_tokens_seen": 7848864, + "step": 6045 + }, + { + "epoch": 0.29560501307013903, + "grad_norm": 0.00380347459577024, + "learning_rate": 0.28338791552350795, + "loss": 0.1198, + "num_input_tokens_seen": 7856064, + "step": 6050 + }, + { + "epoch": 0.29584931473383336, + "grad_norm": 0.005192140582948923, + "learning_rate": 0.28336096120535914, + "loss": 0.1245, + "num_input_tokens_seen": 7862432, + "step": 6055 + }, + { + "epoch": 0.2960936163975277, + "grad_norm": 0.00339495949447155, + "learning_rate": 0.2833339863212741, + "loss": 0.1414, + "num_input_tokens_seen": 7868800, + "step": 6060 + }, + { + "epoch": 0.296337918061222, + "grad_norm": 0.0036287454422563314, + "learning_rate": 0.28330699087541283, + "loss": 0.1429, + "num_input_tokens_seen": 7875040, + "step": 6065 + }, + { + "epoch": 0.29658221972491633, + "grad_norm": 0.0026034715119749308, + "learning_rate": 0.2832799748719384, + "loss": 0.1217, + "num_input_tokens_seen": 7881664, + "step": 6070 + }, + { + "epoch": 0.29682652138861065, + "grad_norm": 0.0019847811199724674, + "learning_rate": 0.28325293831501686, + "loss": 0.1217, + "num_input_tokens_seen": 7888320, + "step": 6075 + }, + { + "epoch": 0.297070823052305, + "grad_norm": 0.0045869797468185425, + "learning_rate": 0.2832258812088177, + "loss": 0.136, + "num_input_tokens_seen": 7894400, + "step": 6080 + }, + { + "epoch": 0.2973151247159993, + "grad_norm": 0.005016650538891554, + "learning_rate": 0.2831988035575134, + "loss": 0.1391, + "num_input_tokens_seen": 7900864, + "step": 6085 + }, + { + "epoch": 0.2975594263796936, + "grad_norm": 0.008143165148794651, + "learning_rate": 0.28317170536527975, + "loss": 0.1522, + "num_input_tokens_seen": 7907488, + "step": 6090 + }, + { + "epoch": 0.29780372804338795, + "grad_norm": 0.005242623388767242, + "learning_rate": 0.2831445866362956, + "loss": 0.1312, + "num_input_tokens_seen": 7913728, + "step": 6095 + }, + { + "epoch": 0.29804802970708233, + "grad_norm": 0.005804647225886583, + "learning_rate": 0.2831174473747429, + "loss": 0.1596, + "num_input_tokens_seen": 7920384, + "step": 6100 + }, + { + "epoch": 0.29829233137077665, + "grad_norm": 0.0027167389634996653, + "learning_rate": 0.2830902875848071, + "loss": 0.1386, + "num_input_tokens_seen": 7927072, + "step": 6105 + }, + { + "epoch": 0.298536633034471, + "grad_norm": 0.003480195067822933, + "learning_rate": 0.28306310727067635, + "loss": 0.1267, + "num_input_tokens_seen": 7933952, + "step": 6110 + }, + { + "epoch": 0.2987809346981653, + "grad_norm": 0.005975958425551653, + "learning_rate": 0.2830359064365423, + "loss": 0.1325, + "num_input_tokens_seen": 7940544, + "step": 6115 + }, + { + "epoch": 0.2990252363618596, + "grad_norm": 0.005594910588115454, + "learning_rate": 0.28300868508659965, + "loss": 0.1273, + "num_input_tokens_seen": 7947040, + "step": 6120 + }, + { + "epoch": 0.29926953802555395, + "grad_norm": 0.004734850954264402, + "learning_rate": 0.28298144322504626, + "loss": 0.1226, + "num_input_tokens_seen": 7953792, + "step": 6125 + }, + { + "epoch": 0.2995138396892483, + "grad_norm": 0.0026224926114082336, + "learning_rate": 0.2829541808560832, + "loss": 0.1293, + "num_input_tokens_seen": 7960448, + "step": 6130 + }, + { + "epoch": 0.2997581413529426, + "grad_norm": 0.0033648270182311535, + "learning_rate": 0.2829268979839146, + "loss": 0.1186, + "num_input_tokens_seen": 7967360, + "step": 6135 + }, + { + "epoch": 0.3000024430166369, + "grad_norm": 0.005862207617610693, + "learning_rate": 0.2828995946127479, + "loss": 0.1555, + "num_input_tokens_seen": 7973664, + "step": 6140 + }, + { + "epoch": 0.30024674468033125, + "grad_norm": 0.0021076686680316925, + "learning_rate": 0.2828722707467936, + "loss": 0.1085, + "num_input_tokens_seen": 7980064, + "step": 6145 + }, + { + "epoch": 0.3004910463440256, + "grad_norm": 0.005318042356520891, + "learning_rate": 0.2828449263902653, + "loss": 0.1286, + "num_input_tokens_seen": 7986720, + "step": 6150 + }, + { + "epoch": 0.30073534800771995, + "grad_norm": 0.0031345009338110685, + "learning_rate": 0.28281756154738, + "loss": 0.1197, + "num_input_tokens_seen": 7992960, + "step": 6155 + }, + { + "epoch": 0.3009796496714143, + "grad_norm": 0.0033173502888530493, + "learning_rate": 0.28279017622235764, + "loss": 0.1149, + "num_input_tokens_seen": 8000544, + "step": 6160 + }, + { + "epoch": 0.3012239513351086, + "grad_norm": 0.0034150085411965847, + "learning_rate": 0.28276277041942127, + "loss": 0.1297, + "num_input_tokens_seen": 8007424, + "step": 6165 + }, + { + "epoch": 0.3014682529988029, + "grad_norm": 0.0030458341352641582, + "learning_rate": 0.2827353441427974, + "loss": 0.1269, + "num_input_tokens_seen": 8013920, + "step": 6170 + }, + { + "epoch": 0.30171255466249725, + "grad_norm": 0.004470922984182835, + "learning_rate": 0.2827078973967153, + "loss": 0.1206, + "num_input_tokens_seen": 8020576, + "step": 6175 + }, + { + "epoch": 0.30195685632619157, + "grad_norm": 0.005556088872253895, + "learning_rate": 0.2826804301854078, + "loss": 0.1788, + "num_input_tokens_seen": 8027008, + "step": 6180 + }, + { + "epoch": 0.3022011579898859, + "grad_norm": 0.003629059996455908, + "learning_rate": 0.2826529425131105, + "loss": 0.1695, + "num_input_tokens_seen": 8033696, + "step": 6185 + }, + { + "epoch": 0.3024454596535802, + "grad_norm": 0.006186873186379671, + "learning_rate": 0.2826254343840625, + "loss": 0.1617, + "num_input_tokens_seen": 8039808, + "step": 6190 + }, + { + "epoch": 0.3026897613172746, + "grad_norm": 0.004068624693900347, + "learning_rate": 0.2825979058025059, + "loss": 0.1383, + "num_input_tokens_seen": 8046208, + "step": 6195 + }, + { + "epoch": 0.3029340629809689, + "grad_norm": 0.002026492962613702, + "learning_rate": 0.2825703567726858, + "loss": 0.1255, + "num_input_tokens_seen": 8052736, + "step": 6200 + }, + { + "epoch": 0.3029340629809689, + "eval_loss": 0.13456138968467712, + "eval_runtime": 402.2055, + "eval_samples_per_second": 90.464, + "eval_steps_per_second": 22.618, + "num_input_tokens_seen": 8052736, + "step": 6200 + }, + { + "epoch": 0.30317836464466325, + "grad_norm": 0.001919546746648848, + "learning_rate": 0.2825427872988508, + "loss": 0.1047, + "num_input_tokens_seen": 8059104, + "step": 6205 + }, + { + "epoch": 0.30342266630835757, + "grad_norm": 0.0028260364197194576, + "learning_rate": 0.28251519738525227, + "loss": 0.1244, + "num_input_tokens_seen": 8065280, + "step": 6210 + }, + { + "epoch": 0.3036669679720519, + "grad_norm": 0.0038642457220703363, + "learning_rate": 0.28248758703614507, + "loss": 0.1257, + "num_input_tokens_seen": 8071808, + "step": 6215 + }, + { + "epoch": 0.3039112696357462, + "grad_norm": 0.002162781311199069, + "learning_rate": 0.28245995625578696, + "loss": 0.1072, + "num_input_tokens_seen": 8078528, + "step": 6220 + }, + { + "epoch": 0.30415557129944054, + "grad_norm": 0.003990956116467714, + "learning_rate": 0.282432305048439, + "loss": 0.1777, + "num_input_tokens_seen": 8084800, + "step": 6225 + }, + { + "epoch": 0.30439987296313487, + "grad_norm": 0.003832306480035186, + "learning_rate": 0.28240463341836536, + "loss": 0.1683, + "num_input_tokens_seen": 8091296, + "step": 6230 + }, + { + "epoch": 0.3046441746268292, + "grad_norm": 0.004154198337346315, + "learning_rate": 0.2823769413698334, + "loss": 0.1495, + "num_input_tokens_seen": 8097376, + "step": 6235 + }, + { + "epoch": 0.3048884762905235, + "grad_norm": 0.003754551289603114, + "learning_rate": 0.2823492289071135, + "loss": 0.146, + "num_input_tokens_seen": 8104128, + "step": 6240 + }, + { + "epoch": 0.3051327779542179, + "grad_norm": 0.002737839240580797, + "learning_rate": 0.2823214960344793, + "loss": 0.1394, + "num_input_tokens_seen": 8110368, + "step": 6245 + }, + { + "epoch": 0.3053770796179122, + "grad_norm": 0.003031394677236676, + "learning_rate": 0.28229374275620756, + "loss": 0.1486, + "num_input_tokens_seen": 8116704, + "step": 6250 + }, + { + "epoch": 0.30562138128160654, + "grad_norm": 0.0026930656749755144, + "learning_rate": 0.28226596907657814, + "loss": 0.1384, + "num_input_tokens_seen": 8123200, + "step": 6255 + }, + { + "epoch": 0.30586568294530087, + "grad_norm": 0.002657659351825714, + "learning_rate": 0.28223817499987414, + "loss": 0.1093, + "num_input_tokens_seen": 8129440, + "step": 6260 + }, + { + "epoch": 0.3061099846089952, + "grad_norm": 0.0036364903207868338, + "learning_rate": 0.2822103605303818, + "loss": 0.1189, + "num_input_tokens_seen": 8135968, + "step": 6265 + }, + { + "epoch": 0.3063542862726895, + "grad_norm": 0.00338316080160439, + "learning_rate": 0.2821825256723903, + "loss": 0.1153, + "num_input_tokens_seen": 8142752, + "step": 6270 + }, + { + "epoch": 0.30659858793638384, + "grad_norm": 0.004045460373163223, + "learning_rate": 0.2821546704301923, + "loss": 0.1274, + "num_input_tokens_seen": 8148928, + "step": 6275 + }, + { + "epoch": 0.30684288960007816, + "grad_norm": 0.004249666351824999, + "learning_rate": 0.2821267948080834, + "loss": 0.1534, + "num_input_tokens_seen": 8155328, + "step": 6280 + }, + { + "epoch": 0.3070871912637725, + "grad_norm": 0.002140228869393468, + "learning_rate": 0.28209889881036226, + "loss": 0.1131, + "num_input_tokens_seen": 8162080, + "step": 6285 + }, + { + "epoch": 0.3073314929274668, + "grad_norm": 0.0032928013242781162, + "learning_rate": 0.28207098244133094, + "loss": 0.1273, + "num_input_tokens_seen": 8169184, + "step": 6290 + }, + { + "epoch": 0.3075757945911612, + "grad_norm": 0.00306359282694757, + "learning_rate": 0.2820430457052943, + "loss": 0.1393, + "num_input_tokens_seen": 8175680, + "step": 6295 + }, + { + "epoch": 0.3078200962548555, + "grad_norm": 0.003448140574619174, + "learning_rate": 0.28201508860656077, + "loss": 0.1319, + "num_input_tokens_seen": 8182496, + "step": 6300 + }, + { + "epoch": 0.30806439791854984, + "grad_norm": 0.003530834335833788, + "learning_rate": 0.2819871111494415, + "loss": 0.1106, + "num_input_tokens_seen": 8189088, + "step": 6305 + }, + { + "epoch": 0.30830869958224416, + "grad_norm": 0.006169646512717009, + "learning_rate": 0.28195911333825113, + "loss": 0.1471, + "num_input_tokens_seen": 8195776, + "step": 6310 + }, + { + "epoch": 0.3085530012459385, + "grad_norm": 0.005479929968714714, + "learning_rate": 0.28193109517730713, + "loss": 0.1377, + "num_input_tokens_seen": 8202272, + "step": 6315 + }, + { + "epoch": 0.3087973029096328, + "grad_norm": 0.006138605996966362, + "learning_rate": 0.2819030566709303, + "loss": 0.1253, + "num_input_tokens_seen": 8208640, + "step": 6320 + }, + { + "epoch": 0.30904160457332713, + "grad_norm": 0.005665651056915522, + "learning_rate": 0.2818749978234445, + "loss": 0.1585, + "num_input_tokens_seen": 8214944, + "step": 6325 + }, + { + "epoch": 0.30928590623702146, + "grad_norm": 0.002486680867150426, + "learning_rate": 0.2818469186391768, + "loss": 0.1068, + "num_input_tokens_seen": 8221696, + "step": 6330 + }, + { + "epoch": 0.3095302079007158, + "grad_norm": 0.0024130044039338827, + "learning_rate": 0.28181881912245743, + "loss": 0.1394, + "num_input_tokens_seen": 8227904, + "step": 6335 + }, + { + "epoch": 0.30977450956441016, + "grad_norm": 0.0032961054239422083, + "learning_rate": 0.2817906992776195, + "loss": 0.1391, + "num_input_tokens_seen": 8234400, + "step": 6340 + }, + { + "epoch": 0.3100188112281045, + "grad_norm": 0.0020452190656214952, + "learning_rate": 0.28176255910899967, + "loss": 0.1269, + "num_input_tokens_seen": 8241088, + "step": 6345 + }, + { + "epoch": 0.3102631128917988, + "grad_norm": 0.0029311918187886477, + "learning_rate": 0.2817343986209373, + "loss": 0.1415, + "num_input_tokens_seen": 8247136, + "step": 6350 + }, + { + "epoch": 0.31050741455549313, + "grad_norm": 0.00214764685370028, + "learning_rate": 0.2817062178177753, + "loss": 0.1407, + "num_input_tokens_seen": 8253280, + "step": 6355 + }, + { + "epoch": 0.31075171621918746, + "grad_norm": 0.005673989187926054, + "learning_rate": 0.2816780167038593, + "loss": 0.1533, + "num_input_tokens_seen": 8259392, + "step": 6360 + }, + { + "epoch": 0.3109960178828818, + "grad_norm": 0.002737842034548521, + "learning_rate": 0.28164979528353834, + "loss": 0.1235, + "num_input_tokens_seen": 8265920, + "step": 6365 + }, + { + "epoch": 0.3112403195465761, + "grad_norm": 0.004126700107008219, + "learning_rate": 0.28162155356116453, + "loss": 0.1431, + "num_input_tokens_seen": 8272960, + "step": 6370 + }, + { + "epoch": 0.31148462121027043, + "grad_norm": 0.0026592391077429056, + "learning_rate": 0.28159329154109314, + "loss": 0.1317, + "num_input_tokens_seen": 8280000, + "step": 6375 + }, + { + "epoch": 0.31172892287396475, + "grad_norm": 0.0027857408858835697, + "learning_rate": 0.28156500922768246, + "loss": 0.1354, + "num_input_tokens_seen": 8286144, + "step": 6380 + }, + { + "epoch": 0.3119732245376591, + "grad_norm": 0.0029285254422575235, + "learning_rate": 0.28153670662529406, + "loss": 0.1501, + "num_input_tokens_seen": 8292320, + "step": 6385 + }, + { + "epoch": 0.31221752620135346, + "grad_norm": 0.003085540607571602, + "learning_rate": 0.28150838373829246, + "loss": 0.1225, + "num_input_tokens_seen": 8298848, + "step": 6390 + }, + { + "epoch": 0.3124618278650478, + "grad_norm": 0.00454979483038187, + "learning_rate": 0.2814800405710455, + "loss": 0.1346, + "num_input_tokens_seen": 8305120, + "step": 6395 + }, + { + "epoch": 0.3127061295287421, + "grad_norm": 0.002960595302283764, + "learning_rate": 0.2814516771279239, + "loss": 0.1161, + "num_input_tokens_seen": 8311808, + "step": 6400 + }, + { + "epoch": 0.3127061295287421, + "eval_loss": 0.12723545730113983, + "eval_runtime": 402.8505, + "eval_samples_per_second": 90.319, + "eval_steps_per_second": 22.582, + "num_input_tokens_seen": 8311808, + "step": 6400 + }, + { + "epoch": 0.31295043119243643, + "grad_norm": 0.0027823520358651876, + "learning_rate": 0.28142329341330186, + "loss": 0.108, + "num_input_tokens_seen": 8317856, + "step": 6405 + }, + { + "epoch": 0.31319473285613075, + "grad_norm": 0.005752666387706995, + "learning_rate": 0.2813948894315564, + "loss": 0.134, + "num_input_tokens_seen": 8324608, + "step": 6410 + }, + { + "epoch": 0.3134390345198251, + "grad_norm": 0.0041052415035665035, + "learning_rate": 0.2813664651870677, + "loss": 0.1222, + "num_input_tokens_seen": 8331456, + "step": 6415 + }, + { + "epoch": 0.3136833361835194, + "grad_norm": 0.005810683127492666, + "learning_rate": 0.28133802068421926, + "loss": 0.1332, + "num_input_tokens_seen": 8337792, + "step": 6420 + }, + { + "epoch": 0.3139276378472137, + "grad_norm": 0.0058490317314863205, + "learning_rate": 0.28130955592739754, + "loss": 0.1222, + "num_input_tokens_seen": 8344032, + "step": 6425 + }, + { + "epoch": 0.31417193951090805, + "grad_norm": 0.003976508043706417, + "learning_rate": 0.2812810709209922, + "loss": 0.1484, + "num_input_tokens_seen": 8350272, + "step": 6430 + }, + { + "epoch": 0.3144162411746024, + "grad_norm": 0.003760404186323285, + "learning_rate": 0.2812525656693959, + "loss": 0.1494, + "num_input_tokens_seen": 8357056, + "step": 6435 + }, + { + "epoch": 0.31466054283829675, + "grad_norm": 0.0017641382291913033, + "learning_rate": 0.28122404017700453, + "loss": 0.1448, + "num_input_tokens_seen": 8363360, + "step": 6440 + }, + { + "epoch": 0.3149048445019911, + "grad_norm": 0.001963495509698987, + "learning_rate": 0.2811954944482171, + "loss": 0.1304, + "num_input_tokens_seen": 8370016, + "step": 6445 + }, + { + "epoch": 0.3151491461656854, + "grad_norm": 0.0031955870799720287, + "learning_rate": 0.2811669284874358, + "loss": 0.1492, + "num_input_tokens_seen": 8376128, + "step": 6450 + }, + { + "epoch": 0.3153934478293797, + "grad_norm": 0.0034956224262714386, + "learning_rate": 0.2811383422990657, + "loss": 0.1596, + "num_input_tokens_seen": 8382176, + "step": 6455 + }, + { + "epoch": 0.31563774949307405, + "grad_norm": 0.002924972213804722, + "learning_rate": 0.2811097358875152, + "loss": 0.1404, + "num_input_tokens_seen": 8388512, + "step": 6460 + }, + { + "epoch": 0.3158820511567684, + "grad_norm": 0.0034850016236305237, + "learning_rate": 0.2810811092571959, + "loss": 0.1431, + "num_input_tokens_seen": 8394816, + "step": 6465 + }, + { + "epoch": 0.3161263528204627, + "grad_norm": 0.004422864411026239, + "learning_rate": 0.28105246241252224, + "loss": 0.1386, + "num_input_tokens_seen": 8400800, + "step": 6470 + }, + { + "epoch": 0.316370654484157, + "grad_norm": 0.0020992918871343136, + "learning_rate": 0.28102379535791194, + "loss": 0.1349, + "num_input_tokens_seen": 8407072, + "step": 6475 + }, + { + "epoch": 0.31661495614785135, + "grad_norm": 0.0027320971712470055, + "learning_rate": 0.2809951080977859, + "loss": 0.1331, + "num_input_tokens_seen": 8413632, + "step": 6480 + }, + { + "epoch": 0.31685925781154567, + "grad_norm": 0.0028902078047394753, + "learning_rate": 0.28096640063656797, + "loss": 0.1209, + "num_input_tokens_seen": 8420128, + "step": 6485 + }, + { + "epoch": 0.31710355947524005, + "grad_norm": 0.003904265584424138, + "learning_rate": 0.2809376729786852, + "loss": 0.1252, + "num_input_tokens_seen": 8426816, + "step": 6490 + }, + { + "epoch": 0.3173478611389344, + "grad_norm": 0.0026544476859271526, + "learning_rate": 0.28090892512856785, + "loss": 0.1446, + "num_input_tokens_seen": 8433088, + "step": 6495 + }, + { + "epoch": 0.3175921628026287, + "grad_norm": 0.0025033163838088512, + "learning_rate": 0.2808801570906491, + "loss": 0.1209, + "num_input_tokens_seen": 8439616, + "step": 6500 + }, + { + "epoch": 0.317836464466323, + "grad_norm": 0.0026304631028324366, + "learning_rate": 0.2808513688693654, + "loss": 0.1072, + "num_input_tokens_seen": 8445952, + "step": 6505 + }, + { + "epoch": 0.31808076613001735, + "grad_norm": 0.002110283123329282, + "learning_rate": 0.28082256046915627, + "loss": 0.1063, + "num_input_tokens_seen": 8451968, + "step": 6510 + }, + { + "epoch": 0.31832506779371167, + "grad_norm": 0.003970382269471884, + "learning_rate": 0.28079373189446427, + "loss": 0.1286, + "num_input_tokens_seen": 8457984, + "step": 6515 + }, + { + "epoch": 0.318569369457406, + "grad_norm": 0.0051743267104029655, + "learning_rate": 0.28076488314973513, + "loss": 0.1499, + "num_input_tokens_seen": 8464288, + "step": 6520 + }, + { + "epoch": 0.3188136711211003, + "grad_norm": 0.0027670501731336117, + "learning_rate": 0.28073601423941774, + "loss": 0.1293, + "num_input_tokens_seen": 8470656, + "step": 6525 + }, + { + "epoch": 0.31905797278479464, + "grad_norm": 0.004071767907589674, + "learning_rate": 0.28070712516796403, + "loss": 0.1314, + "num_input_tokens_seen": 8476672, + "step": 6530 + }, + { + "epoch": 0.319302274448489, + "grad_norm": 0.003067601937800646, + "learning_rate": 0.28067821593982906, + "loss": 0.1106, + "num_input_tokens_seen": 8483168, + "step": 6535 + }, + { + "epoch": 0.31954657611218334, + "grad_norm": 0.0034705260768532753, + "learning_rate": 0.28064928655947097, + "loss": 0.1457, + "num_input_tokens_seen": 8489472, + "step": 6540 + }, + { + "epoch": 0.31979087777587767, + "grad_norm": 0.0034792255610227585, + "learning_rate": 0.28062033703135103, + "loss": 0.1308, + "num_input_tokens_seen": 8496256, + "step": 6545 + }, + { + "epoch": 0.320035179439572, + "grad_norm": 0.006296887062489986, + "learning_rate": 0.2805913673599337, + "loss": 0.1317, + "num_input_tokens_seen": 8503744, + "step": 6550 + }, + { + "epoch": 0.3202794811032663, + "grad_norm": 0.0049893297255039215, + "learning_rate": 0.2805623775496864, + "loss": 0.1358, + "num_input_tokens_seen": 8510560, + "step": 6555 + }, + { + "epoch": 0.32052378276696064, + "grad_norm": 0.002572352299466729, + "learning_rate": 0.2805333676050797, + "loss": 0.1277, + "num_input_tokens_seen": 8517344, + "step": 6560 + }, + { + "epoch": 0.32076808443065497, + "grad_norm": 0.0026679066941142082, + "learning_rate": 0.2805043375305873, + "loss": 0.1385, + "num_input_tokens_seen": 8523616, + "step": 6565 + }, + { + "epoch": 0.3210123860943493, + "grad_norm": 0.004374559968709946, + "learning_rate": 0.2804752873306861, + "loss": 0.1347, + "num_input_tokens_seen": 8529920, + "step": 6570 + }, + { + "epoch": 0.3212566877580436, + "grad_norm": 0.003472002223134041, + "learning_rate": 0.2804462170098559, + "loss": 0.1337, + "num_input_tokens_seen": 8536160, + "step": 6575 + }, + { + "epoch": 0.32150098942173794, + "grad_norm": 0.005676683504134417, + "learning_rate": 0.2804171265725797, + "loss": 0.1204, + "num_input_tokens_seen": 8542464, + "step": 6580 + }, + { + "epoch": 0.3217452910854323, + "grad_norm": 0.0032897561322897673, + "learning_rate": 0.28038801602334373, + "loss": 0.0958, + "num_input_tokens_seen": 8548576, + "step": 6585 + }, + { + "epoch": 0.32198959274912664, + "grad_norm": 0.0029250297229737043, + "learning_rate": 0.28035888536663717, + "loss": 0.1303, + "num_input_tokens_seen": 8555552, + "step": 6590 + }, + { + "epoch": 0.32223389441282096, + "grad_norm": 0.002778823720291257, + "learning_rate": 0.2803297346069522, + "loss": 0.1061, + "num_input_tokens_seen": 8562016, + "step": 6595 + }, + { + "epoch": 0.3224781960765153, + "grad_norm": 0.0030291143339127302, + "learning_rate": 0.28030056374878437, + "loss": 0.1323, + "num_input_tokens_seen": 8568416, + "step": 6600 + }, + { + "epoch": 0.3224781960765153, + "eval_loss": 0.13228586316108704, + "eval_runtime": 402.7949, + "eval_samples_per_second": 90.331, + "eval_steps_per_second": 22.585, + "num_input_tokens_seen": 8568416, + "step": 6600 + }, + { + "epoch": 0.3227224977402096, + "grad_norm": 0.003857929026708007, + "learning_rate": 0.2802713727966321, + "loss": 0.1189, + "num_input_tokens_seen": 8574848, + "step": 6605 + }, + { + "epoch": 0.32296679940390394, + "grad_norm": 0.0035127669107168913, + "learning_rate": 0.28024216175499717, + "loss": 0.1247, + "num_input_tokens_seen": 8581696, + "step": 6610 + }, + { + "epoch": 0.32321110106759826, + "grad_norm": 0.0031965940725058317, + "learning_rate": 0.2802129306283841, + "loss": 0.1301, + "num_input_tokens_seen": 8588480, + "step": 6615 + }, + { + "epoch": 0.3234554027312926, + "grad_norm": 0.0021176966838538647, + "learning_rate": 0.28018367942130074, + "loss": 0.1333, + "num_input_tokens_seen": 8595200, + "step": 6620 + }, + { + "epoch": 0.3236997043949869, + "grad_norm": 0.002701342338696122, + "learning_rate": 0.28015440813825804, + "loss": 0.1397, + "num_input_tokens_seen": 8601824, + "step": 6625 + }, + { + "epoch": 0.32394400605868123, + "grad_norm": 0.0023307071533054113, + "learning_rate": 0.28012511678377006, + "loss": 0.0989, + "num_input_tokens_seen": 8608320, + "step": 6630 + }, + { + "epoch": 0.3241883077223756, + "grad_norm": 0.0023747573141008615, + "learning_rate": 0.28009580536235373, + "loss": 0.093, + "num_input_tokens_seen": 8615552, + "step": 6635 + }, + { + "epoch": 0.32443260938606994, + "grad_norm": 0.004992429167032242, + "learning_rate": 0.28006647387852934, + "loss": 0.1369, + "num_input_tokens_seen": 8622112, + "step": 6640 + }, + { + "epoch": 0.32467691104976426, + "grad_norm": 0.005166053306311369, + "learning_rate": 0.28003712233682015, + "loss": 0.1509, + "num_input_tokens_seen": 8628544, + "step": 6645 + }, + { + "epoch": 0.3249212127134586, + "grad_norm": 0.002472867490723729, + "learning_rate": 0.2800077507417526, + "loss": 0.1089, + "num_input_tokens_seen": 8635200, + "step": 6650 + }, + { + "epoch": 0.3251655143771529, + "grad_norm": 0.001958477543666959, + "learning_rate": 0.2799783590978561, + "loss": 0.1002, + "num_input_tokens_seen": 8641280, + "step": 6655 + }, + { + "epoch": 0.32540981604084723, + "grad_norm": 0.0031052096746861935, + "learning_rate": 0.2799489474096632, + "loss": 0.1456, + "num_input_tokens_seen": 8648160, + "step": 6660 + }, + { + "epoch": 0.32565411770454156, + "grad_norm": 0.004212753847241402, + "learning_rate": 0.27991951568170953, + "loss": 0.1193, + "num_input_tokens_seen": 8654400, + "step": 6665 + }, + { + "epoch": 0.3258984193682359, + "grad_norm": 0.003350746352225542, + "learning_rate": 0.2798900639185339, + "loss": 0.1009, + "num_input_tokens_seen": 8661408, + "step": 6670 + }, + { + "epoch": 0.3261427210319302, + "grad_norm": 0.0028296762611716986, + "learning_rate": 0.2798605921246781, + "loss": 0.1418, + "num_input_tokens_seen": 8667776, + "step": 6675 + }, + { + "epoch": 0.3263870226956246, + "grad_norm": 0.004757254384458065, + "learning_rate": 0.2798311003046871, + "loss": 0.1611, + "num_input_tokens_seen": 8674208, + "step": 6680 + }, + { + "epoch": 0.3266313243593189, + "grad_norm": 0.002534742932766676, + "learning_rate": 0.2798015884631089, + "loss": 0.1199, + "num_input_tokens_seen": 8680576, + "step": 6685 + }, + { + "epoch": 0.32687562602301323, + "grad_norm": 0.0013794016558676958, + "learning_rate": 0.27977205660449445, + "loss": 0.1617, + "num_input_tokens_seen": 8687360, + "step": 6690 + }, + { + "epoch": 0.32711992768670756, + "grad_norm": 0.0025193982291966677, + "learning_rate": 0.2797425047333981, + "loss": 0.1299, + "num_input_tokens_seen": 8693728, + "step": 6695 + }, + { + "epoch": 0.3273642293504019, + "grad_norm": 0.0020046455319970846, + "learning_rate": 0.27971293285437715, + "loss": 0.1552, + "num_input_tokens_seen": 8700160, + "step": 6700 + }, + { + "epoch": 0.3276085310140962, + "grad_norm": 0.004626846872270107, + "learning_rate": 0.2796833409719918, + "loss": 0.1413, + "num_input_tokens_seen": 8706560, + "step": 6705 + }, + { + "epoch": 0.32785283267779053, + "grad_norm": 0.0019206784199923277, + "learning_rate": 0.27965372909080566, + "loss": 0.1299, + "num_input_tokens_seen": 8712672, + "step": 6710 + }, + { + "epoch": 0.32809713434148485, + "grad_norm": 0.0028740086127072573, + "learning_rate": 0.27962409721538506, + "loss": 0.1532, + "num_input_tokens_seen": 8719584, + "step": 6715 + }, + { + "epoch": 0.3283414360051792, + "grad_norm": 0.0023436122573912144, + "learning_rate": 0.27959444535029976, + "loss": 0.1319, + "num_input_tokens_seen": 8726688, + "step": 6720 + }, + { + "epoch": 0.3285857376688735, + "grad_norm": 0.0035289109218865633, + "learning_rate": 0.27956477350012243, + "loss": 0.1383, + "num_input_tokens_seen": 8733440, + "step": 6725 + }, + { + "epoch": 0.3288300393325679, + "grad_norm": 0.002691833768039942, + "learning_rate": 0.27953508166942875, + "loss": 0.1366, + "num_input_tokens_seen": 8739872, + "step": 6730 + }, + { + "epoch": 0.3290743409962622, + "grad_norm": 0.002394879935309291, + "learning_rate": 0.27950536986279767, + "loss": 0.1245, + "num_input_tokens_seen": 8746528, + "step": 6735 + }, + { + "epoch": 0.32931864265995653, + "grad_norm": 0.0027530102524906397, + "learning_rate": 0.2794756380848111, + "loss": 0.1373, + "num_input_tokens_seen": 8752864, + "step": 6740 + }, + { + "epoch": 0.32956294432365085, + "grad_norm": 0.002407361753284931, + "learning_rate": 0.279445886340054, + "loss": 0.1233, + "num_input_tokens_seen": 8759744, + "step": 6745 + }, + { + "epoch": 0.3298072459873452, + "grad_norm": 0.00219430448487401, + "learning_rate": 0.27941611463311455, + "loss": 0.1046, + "num_input_tokens_seen": 8765664, + "step": 6750 + }, + { + "epoch": 0.3300515476510395, + "grad_norm": 0.0032388358376920223, + "learning_rate": 0.2793863229685839, + "loss": 0.1341, + "num_input_tokens_seen": 8771680, + "step": 6755 + }, + { + "epoch": 0.3302958493147338, + "grad_norm": 0.003104954957962036, + "learning_rate": 0.27935651135105627, + "loss": 0.1695, + "num_input_tokens_seen": 8778528, + "step": 6760 + }, + { + "epoch": 0.33054015097842815, + "grad_norm": 0.003269845386967063, + "learning_rate": 0.279326679785129, + "loss": 0.1451, + "num_input_tokens_seen": 8785088, + "step": 6765 + }, + { + "epoch": 0.3307844526421225, + "grad_norm": 0.0025573605671525, + "learning_rate": 0.2792968282754024, + "loss": 0.1182, + "num_input_tokens_seen": 8791360, + "step": 6770 + }, + { + "epoch": 0.3310287543058168, + "grad_norm": 0.0029063464608043432, + "learning_rate": 0.2792669568264801, + "loss": 0.1319, + "num_input_tokens_seen": 8797824, + "step": 6775 + }, + { + "epoch": 0.3312730559695112, + "grad_norm": 0.004732673522084951, + "learning_rate": 0.27923706544296856, + "loss": 0.151, + "num_input_tokens_seen": 8804160, + "step": 6780 + }, + { + "epoch": 0.3315173576332055, + "grad_norm": 0.0016730795614421368, + "learning_rate": 0.2792071541294775, + "loss": 0.1332, + "num_input_tokens_seen": 8810848, + "step": 6785 + }, + { + "epoch": 0.3317616592968998, + "grad_norm": 0.004289437551051378, + "learning_rate": 0.27917722289061947, + "loss": 0.1235, + "num_input_tokens_seen": 8817312, + "step": 6790 + }, + { + "epoch": 0.33200596096059415, + "grad_norm": 0.0034254351630806923, + "learning_rate": 0.27914727173101034, + "loss": 0.1398, + "num_input_tokens_seen": 8824032, + "step": 6795 + }, + { + "epoch": 0.3322502626242885, + "grad_norm": 0.0019132319139316678, + "learning_rate": 0.279117300655269, + "loss": 0.1171, + "num_input_tokens_seen": 8830400, + "step": 6800 + }, + { + "epoch": 0.3322502626242885, + "eval_loss": 0.13483087718486786, + "eval_runtime": 402.1516, + "eval_samples_per_second": 90.476, + "eval_steps_per_second": 22.621, + "num_input_tokens_seen": 8830400, + "step": 6800 + }, + { + "epoch": 0.3324945642879828, + "grad_norm": 0.002074110321700573, + "learning_rate": 0.2790873096680173, + "loss": 0.1275, + "num_input_tokens_seen": 8836864, + "step": 6805 + }, + { + "epoch": 0.3327388659516771, + "grad_norm": 0.0020507483277469873, + "learning_rate": 0.2790572987738802, + "loss": 0.1138, + "num_input_tokens_seen": 8843520, + "step": 6810 + }, + { + "epoch": 0.33298316761537144, + "grad_norm": 0.002114031696692109, + "learning_rate": 0.27902726797748584, + "loss": 0.1137, + "num_input_tokens_seen": 8850016, + "step": 6815 + }, + { + "epoch": 0.33322746927906577, + "grad_norm": 0.004281722009181976, + "learning_rate": 0.2789972172834652, + "loss": 0.1199, + "num_input_tokens_seen": 8856384, + "step": 6820 + }, + { + "epoch": 0.3334717709427601, + "grad_norm": 0.0029327564407140017, + "learning_rate": 0.2789671466964527, + "loss": 0.1227, + "num_input_tokens_seen": 8863072, + "step": 6825 + }, + { + "epoch": 0.3337160726064545, + "grad_norm": 0.0031599407084286213, + "learning_rate": 0.2789370562210854, + "loss": 0.1121, + "num_input_tokens_seen": 8869568, + "step": 6830 + }, + { + "epoch": 0.3339603742701488, + "grad_norm": 0.0016883196076378226, + "learning_rate": 0.27890694586200376, + "loss": 0.1427, + "num_input_tokens_seen": 8876032, + "step": 6835 + }, + { + "epoch": 0.3342046759338431, + "grad_norm": 0.002406889805570245, + "learning_rate": 0.2788768156238511, + "loss": 0.1473, + "num_input_tokens_seen": 8882208, + "step": 6840 + }, + { + "epoch": 0.33444897759753744, + "grad_norm": 0.001954546896740794, + "learning_rate": 0.27884666551127385, + "loss": 0.1336, + "num_input_tokens_seen": 8888608, + "step": 6845 + }, + { + "epoch": 0.33469327926123177, + "grad_norm": 0.003744056448340416, + "learning_rate": 0.2788164955289217, + "loss": 0.1224, + "num_input_tokens_seen": 8895872, + "step": 6850 + }, + { + "epoch": 0.3349375809249261, + "grad_norm": 0.002984281862154603, + "learning_rate": 0.27878630568144697, + "loss": 0.1425, + "num_input_tokens_seen": 8902496, + "step": 6855 + }, + { + "epoch": 0.3351818825886204, + "grad_norm": 0.002982716541737318, + "learning_rate": 0.2787560959735056, + "loss": 0.1342, + "num_input_tokens_seen": 8908832, + "step": 6860 + }, + { + "epoch": 0.33542618425231474, + "grad_norm": 0.0036153621040284634, + "learning_rate": 0.27872586640975616, + "loss": 0.1303, + "num_input_tokens_seen": 8915008, + "step": 6865 + }, + { + "epoch": 0.33567048591600906, + "grad_norm": 0.0032651619985699654, + "learning_rate": 0.27869561699486045, + "loss": 0.133, + "num_input_tokens_seen": 8920992, + "step": 6870 + }, + { + "epoch": 0.33591478757970344, + "grad_norm": 0.002830341923981905, + "learning_rate": 0.2786653477334833, + "loss": 0.1261, + "num_input_tokens_seen": 8927744, + "step": 6875 + }, + { + "epoch": 0.33615908924339777, + "grad_norm": 0.001816625357605517, + "learning_rate": 0.2786350586302926, + "loss": 0.1045, + "num_input_tokens_seen": 8934560, + "step": 6880 + }, + { + "epoch": 0.3364033909070921, + "grad_norm": 0.002336018718779087, + "learning_rate": 0.27860474968995935, + "loss": 0.1283, + "num_input_tokens_seen": 8941344, + "step": 6885 + }, + { + "epoch": 0.3366476925707864, + "grad_norm": 0.0026598304975777864, + "learning_rate": 0.27857442091715756, + "loss": 0.1217, + "num_input_tokens_seen": 8947936, + "step": 6890 + }, + { + "epoch": 0.33689199423448074, + "grad_norm": 0.002947137225419283, + "learning_rate": 0.27854407231656425, + "loss": 0.1284, + "num_input_tokens_seen": 8955136, + "step": 6895 + }, + { + "epoch": 0.33713629589817506, + "grad_norm": 0.004111334681510925, + "learning_rate": 0.2785137038928596, + "loss": 0.1554, + "num_input_tokens_seen": 8961472, + "step": 6900 + }, + { + "epoch": 0.3373805975618694, + "grad_norm": 0.0020397789776325226, + "learning_rate": 0.27848331565072687, + "loss": 0.1106, + "num_input_tokens_seen": 8967808, + "step": 6905 + }, + { + "epoch": 0.3376248992255637, + "grad_norm": 0.0017282790504395962, + "learning_rate": 0.27845290759485225, + "loss": 0.1044, + "num_input_tokens_seen": 8974304, + "step": 6910 + }, + { + "epoch": 0.33786920088925804, + "grad_norm": 0.0027730048168450594, + "learning_rate": 0.278422479729925, + "loss": 0.1153, + "num_input_tokens_seen": 8980928, + "step": 6915 + }, + { + "epoch": 0.33811350255295236, + "grad_norm": 0.0025923382490873337, + "learning_rate": 0.2783920320606375, + "loss": 0.1128, + "num_input_tokens_seen": 8987264, + "step": 6920 + }, + { + "epoch": 0.33835780421664674, + "grad_norm": 0.002699461532756686, + "learning_rate": 0.2783615645916852, + "loss": 0.131, + "num_input_tokens_seen": 8993728, + "step": 6925 + }, + { + "epoch": 0.33860210588034106, + "grad_norm": 0.0019936533644795418, + "learning_rate": 0.2783310773277666, + "loss": 0.1191, + "num_input_tokens_seen": 9000256, + "step": 6930 + }, + { + "epoch": 0.3388464075440354, + "grad_norm": 0.003877925453707576, + "learning_rate": 0.2783005702735831, + "loss": 0.1237, + "num_input_tokens_seen": 9006752, + "step": 6935 + }, + { + "epoch": 0.3390907092077297, + "grad_norm": 0.0024562603794038296, + "learning_rate": 0.2782700434338394, + "loss": 0.1228, + "num_input_tokens_seen": 9013856, + "step": 6940 + }, + { + "epoch": 0.33933501087142404, + "grad_norm": 0.002472254680469632, + "learning_rate": 0.278239496813243, + "loss": 0.1398, + "num_input_tokens_seen": 9020128, + "step": 6945 + }, + { + "epoch": 0.33957931253511836, + "grad_norm": 0.004543050192296505, + "learning_rate": 0.27820893041650463, + "loss": 0.148, + "num_input_tokens_seen": 9026176, + "step": 6950 + }, + { + "epoch": 0.3398236141988127, + "grad_norm": 0.0013973558088764548, + "learning_rate": 0.27817834424833804, + "loss": 0.1156, + "num_input_tokens_seen": 9032576, + "step": 6955 + }, + { + "epoch": 0.340067915862507, + "grad_norm": 0.0033213722053915262, + "learning_rate": 0.27814773831345996, + "loss": 0.1262, + "num_input_tokens_seen": 9038752, + "step": 6960 + }, + { + "epoch": 0.34031221752620133, + "grad_norm": 0.0030946156475692987, + "learning_rate": 0.2781171126165902, + "loss": 0.1174, + "num_input_tokens_seen": 9044960, + "step": 6965 + }, + { + "epoch": 0.34055651918989566, + "grad_norm": 0.00226924242451787, + "learning_rate": 0.2780864671624517, + "loss": 0.1284, + "num_input_tokens_seen": 9051552, + "step": 6970 + }, + { + "epoch": 0.34080082085359004, + "grad_norm": 0.0022124694660305977, + "learning_rate": 0.27805580195577034, + "loss": 0.1384, + "num_input_tokens_seen": 9058048, + "step": 6975 + }, + { + "epoch": 0.34104512251728436, + "grad_norm": 0.0020963980350643396, + "learning_rate": 0.2780251170012751, + "loss": 0.1291, + "num_input_tokens_seen": 9064480, + "step": 6980 + }, + { + "epoch": 0.3412894241809787, + "grad_norm": 0.0023328172974288464, + "learning_rate": 0.27799441230369787, + "loss": 0.1168, + "num_input_tokens_seen": 9071488, + "step": 6985 + }, + { + "epoch": 0.341533725844673, + "grad_norm": 0.0028613926842808723, + "learning_rate": 0.27796368786777387, + "loss": 0.1429, + "num_input_tokens_seen": 9077920, + "step": 6990 + }, + { + "epoch": 0.34177802750836733, + "grad_norm": 0.0030711647123098373, + "learning_rate": 0.277932943698241, + "loss": 0.1192, + "num_input_tokens_seen": 9084672, + "step": 6995 + }, + { + "epoch": 0.34202232917206166, + "grad_norm": 0.0021340837702155113, + "learning_rate": 0.2779021797998406, + "loss": 0.1009, + "num_input_tokens_seen": 9091040, + "step": 7000 + }, + { + "epoch": 0.34202232917206166, + "eval_loss": 0.13825573027133942, + "eval_runtime": 402.6276, + "eval_samples_per_second": 90.369, + "eval_steps_per_second": 22.594, + "num_input_tokens_seen": 9091040, + "step": 7000 + }, + { + "epoch": 0.342266630835756, + "grad_norm": 0.0020070464815944433, + "learning_rate": 0.2778713961773167, + "loss": 0.1252, + "num_input_tokens_seen": 9097312, + "step": 7005 + }, + { + "epoch": 0.3425109324994503, + "grad_norm": 0.002990053268149495, + "learning_rate": 0.2778405928354166, + "loss": 0.1384, + "num_input_tokens_seen": 9104000, + "step": 7010 + }, + { + "epoch": 0.34275523416314463, + "grad_norm": 0.0025197495706379414, + "learning_rate": 0.27780976977889055, + "loss": 0.1132, + "num_input_tokens_seen": 9110304, + "step": 7015 + }, + { + "epoch": 0.342999535826839, + "grad_norm": 0.0026811507996171713, + "learning_rate": 0.27777892701249185, + "loss": 0.1422, + "num_input_tokens_seen": 9116736, + "step": 7020 + }, + { + "epoch": 0.34324383749053333, + "grad_norm": 0.0022911846172064543, + "learning_rate": 0.2777480645409768, + "loss": 0.1351, + "num_input_tokens_seen": 9122912, + "step": 7025 + }, + { + "epoch": 0.34348813915422766, + "grad_norm": 0.0020218209829181433, + "learning_rate": 0.27771718236910486, + "loss": 0.1006, + "num_input_tokens_seen": 9129440, + "step": 7030 + }, + { + "epoch": 0.343732440817922, + "grad_norm": 0.002343345433473587, + "learning_rate": 0.27768628050163835, + "loss": 0.1213, + "num_input_tokens_seen": 9135712, + "step": 7035 + }, + { + "epoch": 0.3439767424816163, + "grad_norm": 0.0020744178909808397, + "learning_rate": 0.2776553589433428, + "loss": 0.1062, + "num_input_tokens_seen": 9142528, + "step": 7040 + }, + { + "epoch": 0.34422104414531063, + "grad_norm": 0.0018253695452585816, + "learning_rate": 0.27762441769898666, + "loss": 0.1249, + "num_input_tokens_seen": 9149472, + "step": 7045 + }, + { + "epoch": 0.34446534580900495, + "grad_norm": 0.002932059345766902, + "learning_rate": 0.2775934567733415, + "loss": 0.1387, + "num_input_tokens_seen": 9155424, + "step": 7050 + }, + { + "epoch": 0.3447096474726993, + "grad_norm": 0.0030944638419896364, + "learning_rate": 0.2775624761711819, + "loss": 0.1149, + "num_input_tokens_seen": 9162048, + "step": 7055 + }, + { + "epoch": 0.3449539491363936, + "grad_norm": 0.0038761920295655727, + "learning_rate": 0.2775314758972854, + "loss": 0.1106, + "num_input_tokens_seen": 9169056, + "step": 7060 + }, + { + "epoch": 0.3451982508000879, + "grad_norm": 0.004873262252658606, + "learning_rate": 0.2775004559564327, + "loss": 0.1146, + "num_input_tokens_seen": 9175424, + "step": 7065 + }, + { + "epoch": 0.3454425524637823, + "grad_norm": 0.005470145959407091, + "learning_rate": 0.2774694163534073, + "loss": 0.1099, + "num_input_tokens_seen": 9181760, + "step": 7070 + }, + { + "epoch": 0.34568685412747663, + "grad_norm": 0.0030831668991595507, + "learning_rate": 0.27743835709299614, + "loss": 0.1373, + "num_input_tokens_seen": 9188832, + "step": 7075 + }, + { + "epoch": 0.34593115579117095, + "grad_norm": 0.0026301895268261433, + "learning_rate": 0.2774072781799888, + "loss": 0.1262, + "num_input_tokens_seen": 9194912, + "step": 7080 + }, + { + "epoch": 0.3461754574548653, + "grad_norm": 0.0018515140982344747, + "learning_rate": 0.27737617961917804, + "loss": 0.1241, + "num_input_tokens_seen": 9201376, + "step": 7085 + }, + { + "epoch": 0.3464197591185596, + "grad_norm": 0.003136067185550928, + "learning_rate": 0.27734506141535964, + "loss": 0.141, + "num_input_tokens_seen": 9207840, + "step": 7090 + }, + { + "epoch": 0.3466640607822539, + "grad_norm": 0.003513040952384472, + "learning_rate": 0.2773139235733325, + "loss": 0.1103, + "num_input_tokens_seen": 9214400, + "step": 7095 + }, + { + "epoch": 0.34690836244594825, + "grad_norm": 0.0034298195969313383, + "learning_rate": 0.2772827660978984, + "loss": 0.1325, + "num_input_tokens_seen": 9220768, + "step": 7100 + }, + { + "epoch": 0.3471526641096426, + "grad_norm": 0.0019972212612628937, + "learning_rate": 0.27725158899386226, + "loss": 0.1359, + "num_input_tokens_seen": 9227296, + "step": 7105 + }, + { + "epoch": 0.3473969657733369, + "grad_norm": 0.0017573405057191849, + "learning_rate": 0.27722039226603196, + "loss": 0.1121, + "num_input_tokens_seen": 9233856, + "step": 7110 + }, + { + "epoch": 0.3476412674370312, + "grad_norm": 0.0026236565317958593, + "learning_rate": 0.2771891759192184, + "loss": 0.1089, + "num_input_tokens_seen": 9239936, + "step": 7115 + }, + { + "epoch": 0.3478855691007256, + "grad_norm": 0.0025627072900533676, + "learning_rate": 0.2771579399582355, + "loss": 0.1193, + "num_input_tokens_seen": 9246048, + "step": 7120 + }, + { + "epoch": 0.3481298707644199, + "grad_norm": 0.001962937880307436, + "learning_rate": 0.2771266843879004, + "loss": 0.1254, + "num_input_tokens_seen": 9252288, + "step": 7125 + }, + { + "epoch": 0.34837417242811425, + "grad_norm": 0.0029408582486212254, + "learning_rate": 0.2770954092130329, + "loss": 0.1164, + "num_input_tokens_seen": 9259520, + "step": 7130 + }, + { + "epoch": 0.3486184740918086, + "grad_norm": 0.003723337547853589, + "learning_rate": 0.27706411443845613, + "loss": 0.1072, + "num_input_tokens_seen": 9265952, + "step": 7135 + }, + { + "epoch": 0.3488627757555029, + "grad_norm": 0.002133642788976431, + "learning_rate": 0.27703280006899617, + "loss": 0.129, + "num_input_tokens_seen": 9272096, + "step": 7140 + }, + { + "epoch": 0.3491070774191972, + "grad_norm": 0.002142610028386116, + "learning_rate": 0.277001466109482, + "loss": 0.1127, + "num_input_tokens_seen": 9278944, + "step": 7145 + }, + { + "epoch": 0.34935137908289154, + "grad_norm": 0.0030007169116288424, + "learning_rate": 0.2769701125647458, + "loss": 0.1229, + "num_input_tokens_seen": 9284992, + "step": 7150 + }, + { + "epoch": 0.34959568074658587, + "grad_norm": 0.002040762919932604, + "learning_rate": 0.27693873943962266, + "loss": 0.139, + "num_input_tokens_seen": 9291520, + "step": 7155 + }, + { + "epoch": 0.3498399824102802, + "grad_norm": 0.0024778342340141535, + "learning_rate": 0.2769073467389506, + "loss": 0.1086, + "num_input_tokens_seen": 9298080, + "step": 7160 + }, + { + "epoch": 0.3500842840739745, + "grad_norm": 0.003192250384017825, + "learning_rate": 0.2768759344675709, + "loss": 0.1473, + "num_input_tokens_seen": 9304192, + "step": 7165 + }, + { + "epoch": 0.3503285857376689, + "grad_norm": 0.002263388829305768, + "learning_rate": 0.27684450263032767, + "loss": 0.1467, + "num_input_tokens_seen": 9310528, + "step": 7170 + }, + { + "epoch": 0.3505728874013632, + "grad_norm": 0.005418370943516493, + "learning_rate": 0.2768130512320682, + "loss": 0.1251, + "num_input_tokens_seen": 9317408, + "step": 7175 + }, + { + "epoch": 0.35081718906505754, + "grad_norm": 0.00346702104434371, + "learning_rate": 0.27678158027764244, + "loss": 0.1232, + "num_input_tokens_seen": 9324000, + "step": 7180 + }, + { + "epoch": 0.35106149072875187, + "grad_norm": 0.0037332396022975445, + "learning_rate": 0.27675008977190385, + "loss": 0.1444, + "num_input_tokens_seen": 9330368, + "step": 7185 + }, + { + "epoch": 0.3513057923924462, + "grad_norm": 0.003154112258926034, + "learning_rate": 0.2767185797197086, + "loss": 0.154, + "num_input_tokens_seen": 9336640, + "step": 7190 + }, + { + "epoch": 0.3515500940561405, + "grad_norm": 0.002642940729856491, + "learning_rate": 0.2766870501259159, + "loss": 0.1243, + "num_input_tokens_seen": 9343392, + "step": 7195 + }, + { + "epoch": 0.35179439571983484, + "grad_norm": 0.002797856228426099, + "learning_rate": 0.276655500995388, + "loss": 0.1267, + "num_input_tokens_seen": 9350272, + "step": 7200 + }, + { + "epoch": 0.35179439571983484, + "eval_loss": 0.12421122193336487, + "eval_runtime": 402.5806, + "eval_samples_per_second": 90.379, + "eval_steps_per_second": 22.597, + "num_input_tokens_seen": 9350272, + "step": 7200 + }, + { + "epoch": 0.35203869738352916, + "grad_norm": 0.0037258495576679707, + "learning_rate": 0.27662393233299015, + "loss": 0.1525, + "num_input_tokens_seen": 9356576, + "step": 7205 + }, + { + "epoch": 0.3522829990472235, + "grad_norm": 0.0012894674437120557, + "learning_rate": 0.27659234414359074, + "loss": 0.1255, + "num_input_tokens_seen": 9362848, + "step": 7210 + }, + { + "epoch": 0.35252730071091787, + "grad_norm": 0.004474330227822065, + "learning_rate": 0.27656073643206097, + "loss": 0.1455, + "num_input_tokens_seen": 9369472, + "step": 7215 + }, + { + "epoch": 0.3527716023746122, + "grad_norm": 0.003992806188762188, + "learning_rate": 0.27652910920327517, + "loss": 0.1515, + "num_input_tokens_seen": 9375360, + "step": 7220 + }, + { + "epoch": 0.3530159040383065, + "grad_norm": 0.0019338426645845175, + "learning_rate": 0.2764974624621107, + "loss": 0.1412, + "num_input_tokens_seen": 9382720, + "step": 7225 + }, + { + "epoch": 0.35326020570200084, + "grad_norm": 0.001319100265391171, + "learning_rate": 0.2764657962134479, + "loss": 0.118, + "num_input_tokens_seen": 9389440, + "step": 7230 + }, + { + "epoch": 0.35350450736569516, + "grad_norm": 0.0017002467066049576, + "learning_rate": 0.27643411046217, + "loss": 0.1486, + "num_input_tokens_seen": 9395520, + "step": 7235 + }, + { + "epoch": 0.3537488090293895, + "grad_norm": 0.0022162024397403, + "learning_rate": 0.27640240521316334, + "loss": 0.1171, + "num_input_tokens_seen": 9401536, + "step": 7240 + }, + { + "epoch": 0.3539931106930838, + "grad_norm": 0.0027015635278075933, + "learning_rate": 0.2763706804713174, + "loss": 0.1262, + "num_input_tokens_seen": 9408320, + "step": 7245 + }, + { + "epoch": 0.35423741235677814, + "grad_norm": 0.006211508996784687, + "learning_rate": 0.2763389362415245, + "loss": 0.1133, + "num_input_tokens_seen": 9414688, + "step": 7250 + }, + { + "epoch": 0.35448171402047246, + "grad_norm": 0.0029202771838754416, + "learning_rate": 0.27630717252867987, + "loss": 0.1071, + "num_input_tokens_seen": 9421536, + "step": 7255 + }, + { + "epoch": 0.3547260156841668, + "grad_norm": 0.004890897776931524, + "learning_rate": 0.276275389337682, + "loss": 0.1233, + "num_input_tokens_seen": 9428032, + "step": 7260 + }, + { + "epoch": 0.35497031734786116, + "grad_norm": 0.0025375005789101124, + "learning_rate": 0.2762435866734322, + "loss": 0.1288, + "num_input_tokens_seen": 9434272, + "step": 7265 + }, + { + "epoch": 0.3552146190115555, + "grad_norm": 0.0022274095099419355, + "learning_rate": 0.27621176454083485, + "loss": 0.1122, + "num_input_tokens_seen": 9440768, + "step": 7270 + }, + { + "epoch": 0.3554589206752498, + "grad_norm": 0.0022265261504799128, + "learning_rate": 0.2761799229447973, + "loss": 0.1148, + "num_input_tokens_seen": 9447328, + "step": 7275 + }, + { + "epoch": 0.35570322233894414, + "grad_norm": 0.00170320610050112, + "learning_rate": 0.27614806189023006, + "loss": 0.1099, + "num_input_tokens_seen": 9454112, + "step": 7280 + }, + { + "epoch": 0.35594752400263846, + "grad_norm": 0.0032757704611867666, + "learning_rate": 0.27611618138204636, + "loss": 0.1362, + "num_input_tokens_seen": 9460224, + "step": 7285 + }, + { + "epoch": 0.3561918256663328, + "grad_norm": 0.003963279537856579, + "learning_rate": 0.2760842814251626, + "loss": 0.1317, + "num_input_tokens_seen": 9466848, + "step": 7290 + }, + { + "epoch": 0.3564361273300271, + "grad_norm": 0.002555412007495761, + "learning_rate": 0.2760523620244982, + "loss": 0.1293, + "num_input_tokens_seen": 9473536, + "step": 7295 + }, + { + "epoch": 0.35668042899372143, + "grad_norm": 0.0031811888329684734, + "learning_rate": 0.27602042318497544, + "loss": 0.1412, + "num_input_tokens_seen": 9480352, + "step": 7300 + }, + { + "epoch": 0.35692473065741576, + "grad_norm": 0.0031037565786391497, + "learning_rate": 0.2759884649115198, + "loss": 0.1267, + "num_input_tokens_seen": 9486656, + "step": 7305 + }, + { + "epoch": 0.3571690323211101, + "grad_norm": 0.0025446456857025623, + "learning_rate": 0.2759564872090596, + "loss": 0.1219, + "num_input_tokens_seen": 9493280, + "step": 7310 + }, + { + "epoch": 0.35741333398480446, + "grad_norm": 0.0013413686538115144, + "learning_rate": 0.2759244900825262, + "loss": 0.1224, + "num_input_tokens_seen": 9499616, + "step": 7315 + }, + { + "epoch": 0.3576576356484988, + "grad_norm": 0.0012052926467731595, + "learning_rate": 0.2758924735368539, + "loss": 0.1031, + "num_input_tokens_seen": 9506432, + "step": 7320 + }, + { + "epoch": 0.3579019373121931, + "grad_norm": 0.0024893563240766525, + "learning_rate": 0.27586043757698014, + "loss": 0.1278, + "num_input_tokens_seen": 9512608, + "step": 7325 + }, + { + "epoch": 0.35814623897588743, + "grad_norm": 0.0016121219377964735, + "learning_rate": 0.27582838220784534, + "loss": 0.1166, + "num_input_tokens_seen": 9519104, + "step": 7330 + }, + { + "epoch": 0.35839054063958176, + "grad_norm": 0.002062853192910552, + "learning_rate": 0.27579630743439265, + "loss": 0.1462, + "num_input_tokens_seen": 9525760, + "step": 7335 + }, + { + "epoch": 0.3586348423032761, + "grad_norm": 0.001982499612495303, + "learning_rate": 0.2757642132615686, + "loss": 0.1301, + "num_input_tokens_seen": 9531616, + "step": 7340 + }, + { + "epoch": 0.3588791439669704, + "grad_norm": 0.0015609990805387497, + "learning_rate": 0.2757320996943223, + "loss": 0.1181, + "num_input_tokens_seen": 9538016, + "step": 7345 + }, + { + "epoch": 0.35912344563066473, + "grad_norm": 0.0017122182762250304, + "learning_rate": 0.2756999667376062, + "loss": 0.1178, + "num_input_tokens_seen": 9544352, + "step": 7350 + }, + { + "epoch": 0.35936774729435905, + "grad_norm": 0.003573786234483123, + "learning_rate": 0.2756678143963756, + "loss": 0.1191, + "num_input_tokens_seen": 9550816, + "step": 7355 + }, + { + "epoch": 0.35961204895805343, + "grad_norm": 0.002491450635716319, + "learning_rate": 0.2756356426755888, + "loss": 0.1112, + "num_input_tokens_seen": 9557152, + "step": 7360 + }, + { + "epoch": 0.35985635062174776, + "grad_norm": 0.0039031612686812878, + "learning_rate": 0.27560345158020705, + "loss": 0.1219, + "num_input_tokens_seen": 9563872, + "step": 7365 + }, + { + "epoch": 0.3601006522854421, + "grad_norm": 0.0012882878072559834, + "learning_rate": 0.27557124111519465, + "loss": 0.1231, + "num_input_tokens_seen": 9570464, + "step": 7370 + }, + { + "epoch": 0.3603449539491364, + "grad_norm": 0.0020206118933856487, + "learning_rate": 0.27553901128551883, + "loss": 0.1201, + "num_input_tokens_seen": 9576864, + "step": 7375 + }, + { + "epoch": 0.36058925561283073, + "grad_norm": 0.003871061373502016, + "learning_rate": 0.2755067620961498, + "loss": 0.1146, + "num_input_tokens_seen": 9583616, + "step": 7380 + }, + { + "epoch": 0.36083355727652505, + "grad_norm": 0.0046318829990923405, + "learning_rate": 0.27547449355206094, + "loss": 0.139, + "num_input_tokens_seen": 9590048, + "step": 7385 + }, + { + "epoch": 0.3610778589402194, + "grad_norm": 0.001114042941480875, + "learning_rate": 0.2754422056582283, + "loss": 0.1177, + "num_input_tokens_seen": 9596480, + "step": 7390 + }, + { + "epoch": 0.3613221606039137, + "grad_norm": 0.001184022519737482, + "learning_rate": 0.27540989841963115, + "loss": 0.1367, + "num_input_tokens_seen": 9602880, + "step": 7395 + }, + { + "epoch": 0.361566462267608, + "grad_norm": 0.0028406954370439053, + "learning_rate": 0.27537757184125167, + "loss": 0.1283, + "num_input_tokens_seen": 9609312, + "step": 7400 + }, + { + "epoch": 0.361566462267608, + "eval_loss": 0.12593118846416473, + "eval_runtime": 402.4141, + "eval_samples_per_second": 90.417, + "eval_steps_per_second": 22.606, + "num_input_tokens_seen": 9609312, + "step": 7400 + }, + { + "epoch": 0.36181076393130235, + "grad_norm": 0.002168842125684023, + "learning_rate": 0.275345225928075, + "loss": 0.1205, + "num_input_tokens_seen": 9615360, + "step": 7405 + }, + { + "epoch": 0.36205506559499673, + "grad_norm": 0.001848980668000877, + "learning_rate": 0.2753128606850893, + "loss": 0.1369, + "num_input_tokens_seen": 9621888, + "step": 7410 + }, + { + "epoch": 0.36229936725869105, + "grad_norm": 0.0023958648089319468, + "learning_rate": 0.2752804761172858, + "loss": 0.1395, + "num_input_tokens_seen": 9628384, + "step": 7415 + }, + { + "epoch": 0.3625436689223854, + "grad_norm": 0.003298250027000904, + "learning_rate": 0.27524807222965836, + "loss": 0.1414, + "num_input_tokens_seen": 9634816, + "step": 7420 + }, + { + "epoch": 0.3627879705860797, + "grad_norm": 0.0025726526509970427, + "learning_rate": 0.27521564902720436, + "loss": 0.1641, + "num_input_tokens_seen": 9640800, + "step": 7425 + }, + { + "epoch": 0.363032272249774, + "grad_norm": 0.002160405507311225, + "learning_rate": 0.2751832065149236, + "loss": 0.1274, + "num_input_tokens_seen": 9646944, + "step": 7430 + }, + { + "epoch": 0.36327657391346835, + "grad_norm": 0.0016923682997003198, + "learning_rate": 0.2751507446978193, + "loss": 0.11, + "num_input_tokens_seen": 9653760, + "step": 7435 + }, + { + "epoch": 0.36352087557716267, + "grad_norm": 0.003916827496141195, + "learning_rate": 0.2751182635808974, + "loss": 0.1215, + "num_input_tokens_seen": 9660480, + "step": 7440 + }, + { + "epoch": 0.363765177240857, + "grad_norm": 0.002783439354971051, + "learning_rate": 0.27508576316916694, + "loss": 0.1048, + "num_input_tokens_seen": 9666752, + "step": 7445 + }, + { + "epoch": 0.3640094789045513, + "grad_norm": 0.002857765182852745, + "learning_rate": 0.2750532434676399, + "loss": 0.1342, + "num_input_tokens_seen": 9673312, + "step": 7450 + }, + { + "epoch": 0.36425378056824564, + "grad_norm": 0.0019760369323194027, + "learning_rate": 0.27502070448133115, + "loss": 0.1084, + "num_input_tokens_seen": 9679872, + "step": 7455 + }, + { + "epoch": 0.36449808223194, + "grad_norm": 0.0021007279865443707, + "learning_rate": 0.2749881462152587, + "loss": 0.1175, + "num_input_tokens_seen": 9686560, + "step": 7460 + }, + { + "epoch": 0.36474238389563435, + "grad_norm": 0.0031044352799654007, + "learning_rate": 0.2749555686744434, + "loss": 0.1382, + "num_input_tokens_seen": 9693056, + "step": 7465 + }, + { + "epoch": 0.36498668555932867, + "grad_norm": 0.0025338432751595974, + "learning_rate": 0.2749229718639091, + "loss": 0.1171, + "num_input_tokens_seen": 9699264, + "step": 7470 + }, + { + "epoch": 0.365230987223023, + "grad_norm": 0.002642212901264429, + "learning_rate": 0.27489035578868265, + "loss": 0.1184, + "num_input_tokens_seen": 9705472, + "step": 7475 + }, + { + "epoch": 0.3654752888867173, + "grad_norm": 0.001850885571911931, + "learning_rate": 0.2748577204537939, + "loss": 0.1039, + "num_input_tokens_seen": 9712672, + "step": 7480 + }, + { + "epoch": 0.36571959055041164, + "grad_norm": 0.002320363186299801, + "learning_rate": 0.2748250658642756, + "loss": 0.1344, + "num_input_tokens_seen": 9719200, + "step": 7485 + }, + { + "epoch": 0.36596389221410597, + "grad_norm": 0.0018324661068618298, + "learning_rate": 0.2747923920251634, + "loss": 0.1184, + "num_input_tokens_seen": 9725696, + "step": 7490 + }, + { + "epoch": 0.3662081938778003, + "grad_norm": 0.0020671100355684757, + "learning_rate": 0.27475969894149627, + "loss": 0.1487, + "num_input_tokens_seen": 9732064, + "step": 7495 + }, + { + "epoch": 0.3664524955414946, + "grad_norm": 0.003149796277284622, + "learning_rate": 0.2747269866183156, + "loss": 0.1344, + "num_input_tokens_seen": 9738432, + "step": 7500 + }, + { + "epoch": 0.36669679720518894, + "grad_norm": 0.0016423652414232492, + "learning_rate": 0.27469425506066625, + "loss": 0.1096, + "num_input_tokens_seen": 9744832, + "step": 7505 + }, + { + "epoch": 0.3669410988688833, + "grad_norm": 0.0013490989804267883, + "learning_rate": 0.27466150427359576, + "loss": 0.1376, + "num_input_tokens_seen": 9751616, + "step": 7510 + }, + { + "epoch": 0.36718540053257764, + "grad_norm": 0.0018916231347247958, + "learning_rate": 0.2746287342621547, + "loss": 0.1154, + "num_input_tokens_seen": 9758176, + "step": 7515 + }, + { + "epoch": 0.36742970219627197, + "grad_norm": 0.002810888923704624, + "learning_rate": 0.2745959450313966, + "loss": 0.134, + "num_input_tokens_seen": 9764672, + "step": 7520 + }, + { + "epoch": 0.3676740038599663, + "grad_norm": 0.0015303079271689057, + "learning_rate": 0.27456313658637804, + "loss": 0.1362, + "num_input_tokens_seen": 9770880, + "step": 7525 + }, + { + "epoch": 0.3679183055236606, + "grad_norm": 0.0012582441559061408, + "learning_rate": 0.27453030893215846, + "loss": 0.1246, + "num_input_tokens_seen": 9777152, + "step": 7530 + }, + { + "epoch": 0.36816260718735494, + "grad_norm": 0.0030383856501430273, + "learning_rate": 0.2744974620738003, + "loss": 0.1219, + "num_input_tokens_seen": 9783680, + "step": 7535 + }, + { + "epoch": 0.36840690885104926, + "grad_norm": 0.0019713121000677347, + "learning_rate": 0.27446459601636897, + "loss": 0.1165, + "num_input_tokens_seen": 9790240, + "step": 7540 + }, + { + "epoch": 0.3686512105147436, + "grad_norm": 0.002882863860577345, + "learning_rate": 0.2744317107649328, + "loss": 0.1401, + "num_input_tokens_seen": 9796576, + "step": 7545 + }, + { + "epoch": 0.3688955121784379, + "grad_norm": 0.0025806433986872435, + "learning_rate": 0.2743988063245631, + "loss": 0.1207, + "num_input_tokens_seen": 9802944, + "step": 7550 + }, + { + "epoch": 0.3691398138421323, + "grad_norm": 0.0023907723370939493, + "learning_rate": 0.2743658827003342, + "loss": 0.1328, + "num_input_tokens_seen": 9809760, + "step": 7555 + }, + { + "epoch": 0.3693841155058266, + "grad_norm": 0.001974572893232107, + "learning_rate": 0.27433293989732327, + "loss": 0.1, + "num_input_tokens_seen": 9816224, + "step": 7560 + }, + { + "epoch": 0.36962841716952094, + "grad_norm": 0.00442466652020812, + "learning_rate": 0.27429997792061056, + "loss": 0.1304, + "num_input_tokens_seen": 9822400, + "step": 7565 + }, + { + "epoch": 0.36987271883321526, + "grad_norm": 0.003477678168565035, + "learning_rate": 0.27426699677527927, + "loss": 0.1528, + "num_input_tokens_seen": 9829248, + "step": 7570 + }, + { + "epoch": 0.3701170204969096, + "grad_norm": 0.0020294233690947294, + "learning_rate": 0.2742339964664154, + "loss": 0.109, + "num_input_tokens_seen": 9835328, + "step": 7575 + }, + { + "epoch": 0.3703613221606039, + "grad_norm": 0.00445043807849288, + "learning_rate": 0.274200976999108, + "loss": 0.1426, + "num_input_tokens_seen": 9841952, + "step": 7580 + }, + { + "epoch": 0.37060562382429824, + "grad_norm": 0.003106245305389166, + "learning_rate": 0.27416793837844916, + "loss": 0.1322, + "num_input_tokens_seen": 9848576, + "step": 7585 + }, + { + "epoch": 0.37084992548799256, + "grad_norm": 0.0021511248778551817, + "learning_rate": 0.27413488060953384, + "loss": 0.1374, + "num_input_tokens_seen": 9854880, + "step": 7590 + }, + { + "epoch": 0.3710942271516869, + "grad_norm": 0.0020999989937990904, + "learning_rate": 0.27410180369745996, + "loss": 0.1348, + "num_input_tokens_seen": 9861088, + "step": 7595 + }, + { + "epoch": 0.3713385288153812, + "grad_norm": 0.003084902884438634, + "learning_rate": 0.27406870764732844, + "loss": 0.1166, + "num_input_tokens_seen": 9867648, + "step": 7600 + }, + { + "epoch": 0.3713385288153812, + "eval_loss": 0.12342973798513412, + "eval_runtime": 401.8699, + "eval_samples_per_second": 90.539, + "eval_steps_per_second": 22.637, + "num_input_tokens_seen": 9867648, + "step": 7600 + }, + { + "epoch": 0.3715828304790756, + "grad_norm": 0.002049222821369767, + "learning_rate": 0.27403559246424297, + "loss": 0.1179, + "num_input_tokens_seen": 9873856, + "step": 7605 + }, + { + "epoch": 0.3718271321427699, + "grad_norm": 0.002550648059695959, + "learning_rate": 0.2740024581533105, + "loss": 0.1302, + "num_input_tokens_seen": 9880352, + "step": 7610 + }, + { + "epoch": 0.37207143380646424, + "grad_norm": 0.002044799504801631, + "learning_rate": 0.2739693047196406, + "loss": 0.1176, + "num_input_tokens_seen": 9886848, + "step": 7615 + }, + { + "epoch": 0.37231573547015856, + "grad_norm": 0.0017144574085250497, + "learning_rate": 0.27393613216834606, + "loss": 0.125, + "num_input_tokens_seen": 9893408, + "step": 7620 + }, + { + "epoch": 0.3725600371338529, + "grad_norm": 0.0018377610249444842, + "learning_rate": 0.2739029405045424, + "loss": 0.14, + "num_input_tokens_seen": 9899968, + "step": 7625 + }, + { + "epoch": 0.3728043387975472, + "grad_norm": 0.0013830572133883834, + "learning_rate": 0.2738697297333483, + "loss": 0.1129, + "num_input_tokens_seen": 9906752, + "step": 7630 + }, + { + "epoch": 0.37304864046124153, + "grad_norm": 0.001437742030248046, + "learning_rate": 0.2738364998598852, + "loss": 0.1038, + "num_input_tokens_seen": 9913632, + "step": 7635 + }, + { + "epoch": 0.37329294212493586, + "grad_norm": 0.0018899746937677264, + "learning_rate": 0.27380325088927765, + "loss": 0.1204, + "num_input_tokens_seen": 9920128, + "step": 7640 + }, + { + "epoch": 0.3735372437886302, + "grad_norm": 0.0019836081191897392, + "learning_rate": 0.27376998282665294, + "loss": 0.1312, + "num_input_tokens_seen": 9926496, + "step": 7645 + }, + { + "epoch": 0.3737815454523245, + "grad_norm": 0.003316993825137615, + "learning_rate": 0.27373669567714154, + "loss": 0.1299, + "num_input_tokens_seen": 9932960, + "step": 7650 + }, + { + "epoch": 0.3740258471160189, + "grad_norm": 0.0021188054233789444, + "learning_rate": 0.27370338944587663, + "loss": 0.1013, + "num_input_tokens_seen": 9939424, + "step": 7655 + }, + { + "epoch": 0.3742701487797132, + "grad_norm": 0.0017606158507987857, + "learning_rate": 0.27367006413799455, + "loss": 0.1105, + "num_input_tokens_seen": 9946400, + "step": 7660 + }, + { + "epoch": 0.37451445044340753, + "grad_norm": 0.002127890009433031, + "learning_rate": 0.2736367197586345, + "loss": 0.1234, + "num_input_tokens_seen": 9952736, + "step": 7665 + }, + { + "epoch": 0.37475875210710186, + "grad_norm": 0.0016976326005533338, + "learning_rate": 0.2736033563129385, + "loss": 0.1148, + "num_input_tokens_seen": 9958816, + "step": 7670 + }, + { + "epoch": 0.3750030537707962, + "grad_norm": 0.001550232875160873, + "learning_rate": 0.27356997380605164, + "loss": 0.0838, + "num_input_tokens_seen": 9965504, + "step": 7675 + }, + { + "epoch": 0.3752473554344905, + "grad_norm": 0.001709447824396193, + "learning_rate": 0.27353657224312194, + "loss": 0.1236, + "num_input_tokens_seen": 9971680, + "step": 7680 + }, + { + "epoch": 0.37549165709818483, + "grad_norm": 0.002377287019044161, + "learning_rate": 0.2735031516293004, + "loss": 0.1459, + "num_input_tokens_seen": 9978080, + "step": 7685 + }, + { + "epoch": 0.37573595876187915, + "grad_norm": 0.002683205297216773, + "learning_rate": 0.2734697119697408, + "loss": 0.1239, + "num_input_tokens_seen": 9984672, + "step": 7690 + }, + { + "epoch": 0.3759802604255735, + "grad_norm": 0.0023123316932469606, + "learning_rate": 0.27343625326959997, + "loss": 0.1024, + "num_input_tokens_seen": 9991136, + "step": 7695 + }, + { + "epoch": 0.37622456208926786, + "grad_norm": 0.002695098053663969, + "learning_rate": 0.27340277553403775, + "loss": 0.1019, + "num_input_tokens_seen": 9997632, + "step": 7700 + }, + { + "epoch": 0.3764688637529622, + "grad_norm": 0.0032365666702389717, + "learning_rate": 0.2733692787682167, + "loss": 0.1309, + "num_input_tokens_seen": 10004544, + "step": 7705 + }, + { + "epoch": 0.3767131654166565, + "grad_norm": 0.00205387850292027, + "learning_rate": 0.27333576297730255, + "loss": 0.11, + "num_input_tokens_seen": 10011200, + "step": 7710 + }, + { + "epoch": 0.3769574670803508, + "grad_norm": 0.002847490832209587, + "learning_rate": 0.2733022281664638, + "loss": 0.145, + "num_input_tokens_seen": 10017504, + "step": 7715 + }, + { + "epoch": 0.37720176874404515, + "grad_norm": 0.002164345933124423, + "learning_rate": 0.273268674340872, + "loss": 0.117, + "num_input_tokens_seen": 10023456, + "step": 7720 + }, + { + "epoch": 0.3774460704077395, + "grad_norm": 0.0023255778942257166, + "learning_rate": 0.27323510150570146, + "loss": 0.09, + "num_input_tokens_seen": 10030464, + "step": 7725 + }, + { + "epoch": 0.3776903720714338, + "grad_norm": 0.001770451315678656, + "learning_rate": 0.27320150966612966, + "loss": 0.0815, + "num_input_tokens_seen": 10037120, + "step": 7730 + }, + { + "epoch": 0.3779346737351281, + "grad_norm": 0.002065513748675585, + "learning_rate": 0.2731678988273368, + "loss": 0.1044, + "num_input_tokens_seen": 10043520, + "step": 7735 + }, + { + "epoch": 0.37817897539882245, + "grad_norm": 0.0013571828603744507, + "learning_rate": 0.27313426899450605, + "loss": 0.0998, + "num_input_tokens_seen": 10049728, + "step": 7740 + }, + { + "epoch": 0.37842327706251677, + "grad_norm": 0.003530866000801325, + "learning_rate": 0.27310062017282366, + "loss": 0.1376, + "num_input_tokens_seen": 10056064, + "step": 7745 + }, + { + "epoch": 0.37866757872621115, + "grad_norm": 0.0037026498466730118, + "learning_rate": 0.2730669523674787, + "loss": 0.1097, + "num_input_tokens_seen": 10062464, + "step": 7750 + }, + { + "epoch": 0.3789118803899055, + "grad_norm": 0.002494010142982006, + "learning_rate": 0.2730332655836631, + "loss": 0.1164, + "num_input_tokens_seen": 10068896, + "step": 7755 + }, + { + "epoch": 0.3791561820535998, + "grad_norm": 0.0017692589899525046, + "learning_rate": 0.2729995598265718, + "loss": 0.1205, + "num_input_tokens_seen": 10075872, + "step": 7760 + }, + { + "epoch": 0.3794004837172941, + "grad_norm": 0.0025179972872138023, + "learning_rate": 0.2729658351014027, + "loss": 0.1131, + "num_input_tokens_seen": 10083104, + "step": 7765 + }, + { + "epoch": 0.37964478538098845, + "grad_norm": 0.0023394147865474224, + "learning_rate": 0.27293209141335656, + "loss": 0.1135, + "num_input_tokens_seen": 10089408, + "step": 7770 + }, + { + "epoch": 0.37988908704468277, + "grad_norm": 0.0016096326289698482, + "learning_rate": 0.27289832876763703, + "loss": 0.1087, + "num_input_tokens_seen": 10095488, + "step": 7775 + }, + { + "epoch": 0.3801333887083771, + "grad_norm": 0.0031759506091475487, + "learning_rate": 0.27286454716945074, + "loss": 0.1005, + "num_input_tokens_seen": 10101856, + "step": 7780 + }, + { + "epoch": 0.3803776903720714, + "grad_norm": 0.0037635029293596745, + "learning_rate": 0.27283074662400725, + "loss": 0.1288, + "num_input_tokens_seen": 10108320, + "step": 7785 + }, + { + "epoch": 0.38062199203576574, + "grad_norm": 0.0024070325307548046, + "learning_rate": 0.2727969271365191, + "loss": 0.1257, + "num_input_tokens_seen": 10114496, + "step": 7790 + }, + { + "epoch": 0.38086629369946007, + "grad_norm": 0.0031683300621807575, + "learning_rate": 0.2727630887122016, + "loss": 0.1151, + "num_input_tokens_seen": 10120768, + "step": 7795 + }, + { + "epoch": 0.38111059536315445, + "grad_norm": 0.004189035389572382, + "learning_rate": 0.27272923135627314, + "loss": 0.1238, + "num_input_tokens_seen": 10127328, + "step": 7800 + }, + { + "epoch": 0.38111059536315445, + "eval_loss": 0.12052765488624573, + "eval_runtime": 401.9088, + "eval_samples_per_second": 90.53, + "eval_steps_per_second": 22.634, + "num_input_tokens_seen": 10127328, + "step": 7800 + }, + { + "epoch": 0.38135489702684877, + "grad_norm": 0.0017912487965077162, + "learning_rate": 0.2726953550739548, + "loss": 0.1353, + "num_input_tokens_seen": 10133760, + "step": 7805 + }, + { + "epoch": 0.3815991986905431, + "grad_norm": 0.003177400678396225, + "learning_rate": 0.27266145987047086, + "loss": 0.1179, + "num_input_tokens_seen": 10140480, + "step": 7810 + }, + { + "epoch": 0.3818435003542374, + "grad_norm": 0.0019356351112946868, + "learning_rate": 0.27262754575104836, + "loss": 0.1219, + "num_input_tokens_seen": 10147264, + "step": 7815 + }, + { + "epoch": 0.38208780201793174, + "grad_norm": 0.002727528801187873, + "learning_rate": 0.27259361272091726, + "loss": 0.1451, + "num_input_tokens_seen": 10153728, + "step": 7820 + }, + { + "epoch": 0.38233210368162607, + "grad_norm": 0.0027826414443552494, + "learning_rate": 0.27255966078531046, + "loss": 0.1241, + "num_input_tokens_seen": 10160288, + "step": 7825 + }, + { + "epoch": 0.3825764053453204, + "grad_norm": 0.0013036481104791164, + "learning_rate": 0.2725256899494638, + "loss": 0.1241, + "num_input_tokens_seen": 10166848, + "step": 7830 + }, + { + "epoch": 0.3828207070090147, + "grad_norm": 0.0029509614687412977, + "learning_rate": 0.272491700218616, + "loss": 0.1016, + "num_input_tokens_seen": 10173632, + "step": 7835 + }, + { + "epoch": 0.38306500867270904, + "grad_norm": 0.0055826581083238125, + "learning_rate": 0.27245769159800876, + "loss": 0.1461, + "num_input_tokens_seen": 10180096, + "step": 7840 + }, + { + "epoch": 0.38330931033640336, + "grad_norm": 0.0023483007680624723, + "learning_rate": 0.2724236640928865, + "loss": 0.1397, + "num_input_tokens_seen": 10186528, + "step": 7845 + }, + { + "epoch": 0.38355361200009774, + "grad_norm": 0.0012781292898580432, + "learning_rate": 0.27238961770849673, + "loss": 0.1052, + "num_input_tokens_seen": 10192992, + "step": 7850 + }, + { + "epoch": 0.38379791366379207, + "grad_norm": 0.0013371494133025408, + "learning_rate": 0.27235555245008997, + "loss": 0.0995, + "num_input_tokens_seen": 10198784, + "step": 7855 + }, + { + "epoch": 0.3840422153274864, + "grad_norm": 0.001777885016053915, + "learning_rate": 0.2723214683229193, + "loss": 0.1269, + "num_input_tokens_seen": 10205344, + "step": 7860 + }, + { + "epoch": 0.3842865169911807, + "grad_norm": 0.0031794551759958267, + "learning_rate": 0.27228736533224107, + "loss": 0.1604, + "num_input_tokens_seen": 10211840, + "step": 7865 + }, + { + "epoch": 0.38453081865487504, + "grad_norm": 0.001560469507239759, + "learning_rate": 0.27225324348331437, + "loss": 0.1242, + "num_input_tokens_seen": 10218496, + "step": 7870 + }, + { + "epoch": 0.38477512031856936, + "grad_norm": 0.0014136170502752066, + "learning_rate": 0.27221910278140116, + "loss": 0.1044, + "num_input_tokens_seen": 10225248, + "step": 7875 + }, + { + "epoch": 0.3850194219822637, + "grad_norm": 0.0013797241263091564, + "learning_rate": 0.2721849432317664, + "loss": 0.149, + "num_input_tokens_seen": 10231808, + "step": 7880 + }, + { + "epoch": 0.385263723645958, + "grad_norm": 0.001923521631397307, + "learning_rate": 0.2721507648396779, + "loss": 0.0958, + "num_input_tokens_seen": 10238176, + "step": 7885 + }, + { + "epoch": 0.38550802530965234, + "grad_norm": 0.002184925600886345, + "learning_rate": 0.27211656761040653, + "loss": 0.1306, + "num_input_tokens_seen": 10244704, + "step": 7890 + }, + { + "epoch": 0.3857523269733467, + "grad_norm": 0.0019107665866613388, + "learning_rate": 0.2720823515492257, + "loss": 0.1347, + "num_input_tokens_seen": 10250688, + "step": 7895 + }, + { + "epoch": 0.38599662863704104, + "grad_norm": 0.0020972401835024357, + "learning_rate": 0.27204811666141215, + "loss": 0.1298, + "num_input_tokens_seen": 10257088, + "step": 7900 + }, + { + "epoch": 0.38624093030073536, + "grad_norm": 0.0020473720505833626, + "learning_rate": 0.2720138629522452, + "loss": 0.1418, + "num_input_tokens_seen": 10263488, + "step": 7905 + }, + { + "epoch": 0.3864852319644297, + "grad_norm": 0.004475435707718134, + "learning_rate": 0.2719795904270073, + "loss": 0.1203, + "num_input_tokens_seen": 10269984, + "step": 7910 + }, + { + "epoch": 0.386729533628124, + "grad_norm": 0.002370283240452409, + "learning_rate": 0.2719452990909837, + "loss": 0.1135, + "num_input_tokens_seen": 10276576, + "step": 7915 + }, + { + "epoch": 0.38697383529181834, + "grad_norm": 0.001713331090286374, + "learning_rate": 0.2719109889494625, + "loss": 0.1115, + "num_input_tokens_seen": 10282656, + "step": 7920 + }, + { + "epoch": 0.38721813695551266, + "grad_norm": 0.003303506877273321, + "learning_rate": 0.27187666000773475, + "loss": 0.13, + "num_input_tokens_seen": 10288992, + "step": 7925 + }, + { + "epoch": 0.387462438619207, + "grad_norm": 0.0024179660249501467, + "learning_rate": 0.2718423122710944, + "loss": 0.1332, + "num_input_tokens_seen": 10295232, + "step": 7930 + }, + { + "epoch": 0.3877067402829013, + "grad_norm": 0.003044150536879897, + "learning_rate": 0.2718079457448384, + "loss": 0.1094, + "num_input_tokens_seen": 10301920, + "step": 7935 + }, + { + "epoch": 0.38795104194659563, + "grad_norm": 0.0022416480351239443, + "learning_rate": 0.27177356043426637, + "loss": 0.1225, + "num_input_tokens_seen": 10307712, + "step": 7940 + }, + { + "epoch": 0.38819534361029, + "grad_norm": 0.004101255442947149, + "learning_rate": 0.27173915634468104, + "loss": 0.1256, + "num_input_tokens_seen": 10314112, + "step": 7945 + }, + { + "epoch": 0.38843964527398434, + "grad_norm": 0.0017383542144671082, + "learning_rate": 0.27170473348138796, + "loss": 0.1038, + "num_input_tokens_seen": 10320608, + "step": 7950 + }, + { + "epoch": 0.38868394693767866, + "grad_norm": 0.0025607720017433167, + "learning_rate": 0.27167029184969554, + "loss": 0.1355, + "num_input_tokens_seen": 10326848, + "step": 7955 + }, + { + "epoch": 0.388928248601373, + "grad_norm": 0.0018131325487047434, + "learning_rate": 0.27163583145491504, + "loss": 0.1019, + "num_input_tokens_seen": 10332864, + "step": 7960 + }, + { + "epoch": 0.3891725502650673, + "grad_norm": 0.0018715865444391966, + "learning_rate": 0.2716013523023608, + "loss": 0.1287, + "num_input_tokens_seen": 10339008, + "step": 7965 + }, + { + "epoch": 0.38941685192876163, + "grad_norm": 0.0015323760453611612, + "learning_rate": 0.27156685439734995, + "loss": 0.125, + "num_input_tokens_seen": 10345504, + "step": 7970 + }, + { + "epoch": 0.38966115359245596, + "grad_norm": 0.001849826192483306, + "learning_rate": 0.2715323377452024, + "loss": 0.1256, + "num_input_tokens_seen": 10352064, + "step": 7975 + }, + { + "epoch": 0.3899054552561503, + "grad_norm": 0.002213564934208989, + "learning_rate": 0.2714978023512411, + "loss": 0.1312, + "num_input_tokens_seen": 10358496, + "step": 7980 + }, + { + "epoch": 0.3901497569198446, + "grad_norm": 0.0031396602280437946, + "learning_rate": 0.2714632482207918, + "loss": 0.1243, + "num_input_tokens_seen": 10364864, + "step": 7985 + }, + { + "epoch": 0.3903940585835389, + "grad_norm": 0.0021937869023531675, + "learning_rate": 0.2714286753591833, + "loss": 0.1082, + "num_input_tokens_seen": 10371264, + "step": 7990 + }, + { + "epoch": 0.3906383602472333, + "grad_norm": 0.0027515850961208344, + "learning_rate": 0.27139408377174706, + "loss": 0.1321, + "num_input_tokens_seen": 10377568, + "step": 7995 + }, + { + "epoch": 0.39088266191092763, + "grad_norm": 0.0016413168050348759, + "learning_rate": 0.27135947346381756, + "loss": 0.1282, + "num_input_tokens_seen": 10383808, + "step": 8000 + }, + { + "epoch": 0.39088266191092763, + "eval_loss": 0.12104450166225433, + "eval_runtime": 402.7819, + "eval_samples_per_second": 90.334, + "eval_steps_per_second": 22.585, + "num_input_tokens_seen": 10383808, + "step": 8000 + }, + { + "epoch": 0.39112696357462196, + "grad_norm": 0.0020676685962826014, + "learning_rate": 0.2713248444407322, + "loss": 0.1369, + "num_input_tokens_seen": 10390048, + "step": 8005 + }, + { + "epoch": 0.3913712652383163, + "grad_norm": 0.0018032853258773685, + "learning_rate": 0.27129019670783106, + "loss": 0.101, + "num_input_tokens_seen": 10397248, + "step": 8010 + }, + { + "epoch": 0.3916155669020106, + "grad_norm": 0.0027258743066340685, + "learning_rate": 0.27125553027045746, + "loss": 0.1234, + "num_input_tokens_seen": 10404224, + "step": 8015 + }, + { + "epoch": 0.3918598685657049, + "grad_norm": 0.0026562896091490984, + "learning_rate": 0.2712208451339572, + "loss": 0.1323, + "num_input_tokens_seen": 10410272, + "step": 8020 + }, + { + "epoch": 0.39210417022939925, + "grad_norm": 0.002526727970689535, + "learning_rate": 0.27118614130367935, + "loss": 0.1298, + "num_input_tokens_seen": 10416320, + "step": 8025 + }, + { + "epoch": 0.3923484718930936, + "grad_norm": 0.002498963614925742, + "learning_rate": 0.2711514187849756, + "loss": 0.0959, + "num_input_tokens_seen": 10422912, + "step": 8030 + }, + { + "epoch": 0.3925927735567879, + "grad_norm": 0.002390380948781967, + "learning_rate": 0.27111667758320057, + "loss": 0.1146, + "num_input_tokens_seen": 10429664, + "step": 8035 + }, + { + "epoch": 0.3928370752204823, + "grad_norm": 0.0020587944891303778, + "learning_rate": 0.27108191770371176, + "loss": 0.1125, + "num_input_tokens_seen": 10436224, + "step": 8040 + }, + { + "epoch": 0.3930813768841766, + "grad_norm": 0.003427353221923113, + "learning_rate": 0.2710471391518697, + "loss": 0.1474, + "num_input_tokens_seen": 10443168, + "step": 8045 + }, + { + "epoch": 0.3933256785478709, + "grad_norm": 0.0015256068436428905, + "learning_rate": 0.2710123419330375, + "loss": 0.1032, + "num_input_tokens_seen": 10449600, + "step": 8050 + }, + { + "epoch": 0.39356998021156525, + "grad_norm": 0.0013021731283515692, + "learning_rate": 0.2709775260525816, + "loss": 0.1197, + "num_input_tokens_seen": 10455968, + "step": 8055 + }, + { + "epoch": 0.3938142818752596, + "grad_norm": 0.0019237123196944594, + "learning_rate": 0.27094269151587075, + "loss": 0.1031, + "num_input_tokens_seen": 10462592, + "step": 8060 + }, + { + "epoch": 0.3940585835389539, + "grad_norm": 0.0025090433191508055, + "learning_rate": 0.27090783832827703, + "loss": 0.113, + "num_input_tokens_seen": 10469024, + "step": 8065 + }, + { + "epoch": 0.3943028852026482, + "grad_norm": 0.002161596203222871, + "learning_rate": 0.2708729664951753, + "loss": 0.1002, + "num_input_tokens_seen": 10476320, + "step": 8070 + }, + { + "epoch": 0.39454718686634255, + "grad_norm": 0.0038104259874671698, + "learning_rate": 0.27083807602194304, + "loss": 0.1148, + "num_input_tokens_seen": 10482976, + "step": 8075 + }, + { + "epoch": 0.39479148853003687, + "grad_norm": 0.002516265958547592, + "learning_rate": 0.270803166913961, + "loss": 0.1013, + "num_input_tokens_seen": 10489632, + "step": 8080 + }, + { + "epoch": 0.3950357901937312, + "grad_norm": 0.0021680137142539024, + "learning_rate": 0.27076823917661247, + "loss": 0.1036, + "num_input_tokens_seen": 10496000, + "step": 8085 + }, + { + "epoch": 0.3952800918574256, + "grad_norm": 0.0038401037454605103, + "learning_rate": 0.2707332928152838, + "loss": 0.1085, + "num_input_tokens_seen": 10502208, + "step": 8090 + }, + { + "epoch": 0.3955243935211199, + "grad_norm": 0.0021904471796005964, + "learning_rate": 0.2706983278353641, + "loss": 0.1116, + "num_input_tokens_seen": 10508544, + "step": 8095 + }, + { + "epoch": 0.3957686951848142, + "grad_norm": 0.00478410767391324, + "learning_rate": 0.27066334424224553, + "loss": 0.1461, + "num_input_tokens_seen": 10514752, + "step": 8100 + }, + { + "epoch": 0.39601299684850855, + "grad_norm": 0.0024337219074368477, + "learning_rate": 0.27062834204132297, + "loss": 0.1443, + "num_input_tokens_seen": 10520992, + "step": 8105 + }, + { + "epoch": 0.39625729851220287, + "grad_norm": 0.0015252643497660756, + "learning_rate": 0.27059332123799407, + "loss": 0.1535, + "num_input_tokens_seen": 10527456, + "step": 8110 + }, + { + "epoch": 0.3965016001758972, + "grad_norm": 0.001838701544329524, + "learning_rate": 0.27055828183765956, + "loss": 0.1079, + "num_input_tokens_seen": 10534080, + "step": 8115 + }, + { + "epoch": 0.3967459018395915, + "grad_norm": 0.0026454380713403225, + "learning_rate": 0.270523223845723, + "loss": 0.1158, + "num_input_tokens_seen": 10540576, + "step": 8120 + }, + { + "epoch": 0.39699020350328584, + "grad_norm": 0.0025884921196848154, + "learning_rate": 0.2704881472675907, + "loss": 0.1267, + "num_input_tokens_seen": 10547392, + "step": 8125 + }, + { + "epoch": 0.39723450516698017, + "grad_norm": 0.0027118995785713196, + "learning_rate": 0.270453052108672, + "loss": 0.0975, + "num_input_tokens_seen": 10553632, + "step": 8130 + }, + { + "epoch": 0.3974788068306745, + "grad_norm": 0.002633922267705202, + "learning_rate": 0.2704179383743789, + "loss": 0.1137, + "num_input_tokens_seen": 10560064, + "step": 8135 + }, + { + "epoch": 0.39772310849436887, + "grad_norm": 0.002194406231865287, + "learning_rate": 0.27038280607012644, + "loss": 0.1246, + "num_input_tokens_seen": 10566944, + "step": 8140 + }, + { + "epoch": 0.3979674101580632, + "grad_norm": 0.002468876773491502, + "learning_rate": 0.27034765520133247, + "loss": 0.1083, + "num_input_tokens_seen": 10572992, + "step": 8145 + }, + { + "epoch": 0.3982117118217575, + "grad_norm": 0.001767450012266636, + "learning_rate": 0.2703124857734177, + "loss": 0.1376, + "num_input_tokens_seen": 10579104, + "step": 8150 + }, + { + "epoch": 0.39845601348545184, + "grad_norm": 0.0026503901463001966, + "learning_rate": 0.27027729779180565, + "loss": 0.1059, + "num_input_tokens_seen": 10585248, + "step": 8155 + }, + { + "epoch": 0.39870031514914617, + "grad_norm": 0.0027843150310218334, + "learning_rate": 0.27024209126192283, + "loss": 0.1273, + "num_input_tokens_seen": 10592256, + "step": 8160 + }, + { + "epoch": 0.3989446168128405, + "grad_norm": 0.002124324906617403, + "learning_rate": 0.2702068661891984, + "loss": 0.1274, + "num_input_tokens_seen": 10598720, + "step": 8165 + }, + { + "epoch": 0.3991889184765348, + "grad_norm": 0.0013394440757110715, + "learning_rate": 0.2701716225790647, + "loss": 0.0995, + "num_input_tokens_seen": 10605856, + "step": 8170 + }, + { + "epoch": 0.39943322014022914, + "grad_norm": 0.0027202204801142216, + "learning_rate": 0.27013636043695655, + "loss": 0.1091, + "num_input_tokens_seen": 10612256, + "step": 8175 + }, + { + "epoch": 0.39967752180392346, + "grad_norm": 0.004821546841412783, + "learning_rate": 0.27010107976831194, + "loss": 0.1568, + "num_input_tokens_seen": 10618144, + "step": 8180 + }, + { + "epoch": 0.3999218234676178, + "grad_norm": 0.0021518918219953775, + "learning_rate": 0.2700657805785715, + "loss": 0.1109, + "num_input_tokens_seen": 10624192, + "step": 8185 + }, + { + "epoch": 0.40016612513131217, + "grad_norm": 0.0015377324307337403, + "learning_rate": 0.2700304628731789, + "loss": 0.0945, + "num_input_tokens_seen": 10630816, + "step": 8190 + }, + { + "epoch": 0.4004104267950065, + "grad_norm": 0.002189243445172906, + "learning_rate": 0.26999512665758046, + "loss": 0.1266, + "num_input_tokens_seen": 10637088, + "step": 8195 + }, + { + "epoch": 0.4006547284587008, + "grad_norm": 0.0031488381791859865, + "learning_rate": 0.2699597719372256, + "loss": 0.1402, + "num_input_tokens_seen": 10643424, + "step": 8200 + }, + { + "epoch": 0.4006547284587008, + "eval_loss": 0.1194489598274231, + "eval_runtime": 402.2335, + "eval_samples_per_second": 90.457, + "eval_steps_per_second": 22.616, + "num_input_tokens_seen": 10643424, + "step": 8200 + }, + { + "epoch": 0.40089903012239514, + "grad_norm": 0.0015146147925406694, + "learning_rate": 0.26992439871756635, + "loss": 0.0973, + "num_input_tokens_seen": 10650144, + "step": 8205 + }, + { + "epoch": 0.40114333178608946, + "grad_norm": 0.002002829685807228, + "learning_rate": 0.2698890070040578, + "loss": 0.1213, + "num_input_tokens_seen": 10656544, + "step": 8210 + }, + { + "epoch": 0.4013876334497838, + "grad_norm": 0.0029545146971940994, + "learning_rate": 0.2698535968021577, + "loss": 0.1207, + "num_input_tokens_seen": 10663040, + "step": 8215 + }, + { + "epoch": 0.4016319351134781, + "grad_norm": 0.002086552092805505, + "learning_rate": 0.26981816811732684, + "loss": 0.1207, + "num_input_tokens_seen": 10669280, + "step": 8220 + }, + { + "epoch": 0.40187623677717244, + "grad_norm": 0.003230522619560361, + "learning_rate": 0.26978272095502875, + "loss": 0.0993, + "num_input_tokens_seen": 10675552, + "step": 8225 + }, + { + "epoch": 0.40212053844086676, + "grad_norm": 0.003049519145861268, + "learning_rate": 0.26974725532072974, + "loss": 0.1417, + "num_input_tokens_seen": 10681792, + "step": 8230 + }, + { + "epoch": 0.40236484010456114, + "grad_norm": 0.0014684071065858006, + "learning_rate": 0.26971177121989914, + "loss": 0.114, + "num_input_tokens_seen": 10688320, + "step": 8235 + }, + { + "epoch": 0.40260914176825546, + "grad_norm": 0.0020681116729974747, + "learning_rate": 0.2696762686580091, + "loss": 0.141, + "num_input_tokens_seen": 10694112, + "step": 8240 + }, + { + "epoch": 0.4028534434319498, + "grad_norm": 0.002046123845502734, + "learning_rate": 0.26964074764053436, + "loss": 0.1178, + "num_input_tokens_seen": 10700288, + "step": 8245 + }, + { + "epoch": 0.4030977450956441, + "grad_norm": 0.0015810613986104727, + "learning_rate": 0.2696052081729529, + "loss": 0.1082, + "num_input_tokens_seen": 10706720, + "step": 8250 + }, + { + "epoch": 0.40334204675933844, + "grad_norm": 0.002118562115356326, + "learning_rate": 0.2695696502607453, + "loss": 0.1016, + "num_input_tokens_seen": 10713440, + "step": 8255 + }, + { + "epoch": 0.40358634842303276, + "grad_norm": 0.002548815682530403, + "learning_rate": 0.26953407390939504, + "loss": 0.1374, + "num_input_tokens_seen": 10720128, + "step": 8260 + }, + { + "epoch": 0.4038306500867271, + "grad_norm": 0.0017595704412087798, + "learning_rate": 0.26949847912438835, + "loss": 0.0946, + "num_input_tokens_seen": 10726272, + "step": 8265 + }, + { + "epoch": 0.4040749517504214, + "grad_norm": 0.0017366640968248248, + "learning_rate": 0.26946286591121454, + "loss": 0.1187, + "num_input_tokens_seen": 10732192, + "step": 8270 + }, + { + "epoch": 0.40431925341411573, + "grad_norm": 0.001920292736031115, + "learning_rate": 0.2694272342753655, + "loss": 0.1013, + "num_input_tokens_seen": 10738272, + "step": 8275 + }, + { + "epoch": 0.40456355507781006, + "grad_norm": 0.001016492722555995, + "learning_rate": 0.26939158422233617, + "loss": 0.0934, + "num_input_tokens_seen": 10745088, + "step": 8280 + }, + { + "epoch": 0.40480785674150443, + "grad_norm": 0.0018402293790131807, + "learning_rate": 0.26935591575762413, + "loss": 0.1184, + "num_input_tokens_seen": 10751456, + "step": 8285 + }, + { + "epoch": 0.40505215840519876, + "grad_norm": 0.00346456584520638, + "learning_rate": 0.26932022888672996, + "loss": 0.1518, + "num_input_tokens_seen": 10758016, + "step": 8290 + }, + { + "epoch": 0.4052964600688931, + "grad_norm": 0.002272997749969363, + "learning_rate": 0.26928452361515703, + "loss": 0.1083, + "num_input_tokens_seen": 10764064, + "step": 8295 + }, + { + "epoch": 0.4055407617325874, + "grad_norm": 0.0017569845076650381, + "learning_rate": 0.26924879994841155, + "loss": 0.1509, + "num_input_tokens_seen": 10770400, + "step": 8300 + }, + { + "epoch": 0.40578506339628173, + "grad_norm": 0.0019188195001333952, + "learning_rate": 0.2692130578920025, + "loss": 0.1226, + "num_input_tokens_seen": 10777312, + "step": 8305 + }, + { + "epoch": 0.40602936505997606, + "grad_norm": 0.0031161285005509853, + "learning_rate": 0.26917729745144187, + "loss": 0.1285, + "num_input_tokens_seen": 10783840, + "step": 8310 + }, + { + "epoch": 0.4062736667236704, + "grad_norm": 0.0011725384974852204, + "learning_rate": 0.2691415186322443, + "loss": 0.1069, + "num_input_tokens_seen": 10790272, + "step": 8315 + }, + { + "epoch": 0.4065179683873647, + "grad_norm": 0.0026635429821908474, + "learning_rate": 0.2691057214399273, + "loss": 0.1231, + "num_input_tokens_seen": 10796320, + "step": 8320 + }, + { + "epoch": 0.406762270051059, + "grad_norm": 0.0012246649712324142, + "learning_rate": 0.2690699058800113, + "loss": 0.1036, + "num_input_tokens_seen": 10803072, + "step": 8325 + }, + { + "epoch": 0.40700657171475335, + "grad_norm": 0.0015835892409086227, + "learning_rate": 0.2690340719580194, + "loss": 0.1066, + "num_input_tokens_seen": 10809856, + "step": 8330 + }, + { + "epoch": 0.40725087337844773, + "grad_norm": 0.002570734592154622, + "learning_rate": 0.2689982196794778, + "loss": 0.1337, + "num_input_tokens_seen": 10816064, + "step": 8335 + }, + { + "epoch": 0.40749517504214205, + "grad_norm": 0.0028382979799062014, + "learning_rate": 0.2689623490499153, + "loss": 0.1306, + "num_input_tokens_seen": 10822976, + "step": 8340 + }, + { + "epoch": 0.4077394767058364, + "grad_norm": 0.0019881308544427156, + "learning_rate": 0.2689264600748636, + "loss": 0.1011, + "num_input_tokens_seen": 10829312, + "step": 8345 + }, + { + "epoch": 0.4079837783695307, + "grad_norm": 0.00104097172152251, + "learning_rate": 0.26889055275985724, + "loss": 0.0994, + "num_input_tokens_seen": 10835584, + "step": 8350 + }, + { + "epoch": 0.408228080033225, + "grad_norm": 0.0011896246578544378, + "learning_rate": 0.2688546271104335, + "loss": 0.0947, + "num_input_tokens_seen": 10842272, + "step": 8355 + }, + { + "epoch": 0.40847238169691935, + "grad_norm": 0.001988227479159832, + "learning_rate": 0.26881868313213275, + "loss": 0.1181, + "num_input_tokens_seen": 10848576, + "step": 8360 + }, + { + "epoch": 0.4087166833606137, + "grad_norm": 0.0028741061687469482, + "learning_rate": 0.2687827208304978, + "loss": 0.1311, + "num_input_tokens_seen": 10856064, + "step": 8365 + }, + { + "epoch": 0.408960985024308, + "grad_norm": 0.0017674355767667294, + "learning_rate": 0.26874674021107464, + "loss": 0.112, + "num_input_tokens_seen": 10863008, + "step": 8370 + }, + { + "epoch": 0.4092052866880023, + "grad_norm": 0.0017061857506632805, + "learning_rate": 0.2687107412794118, + "loss": 0.0909, + "num_input_tokens_seen": 10869568, + "step": 8375 + }, + { + "epoch": 0.4094495883516967, + "grad_norm": 0.0016132640885189176, + "learning_rate": 0.26867472404106096, + "loss": 0.0895, + "num_input_tokens_seen": 10876064, + "step": 8380 + }, + { + "epoch": 0.409693890015391, + "grad_norm": 0.003455166006460786, + "learning_rate": 0.26863868850157624, + "loss": 0.0763, + "num_input_tokens_seen": 10882656, + "step": 8385 + }, + { + "epoch": 0.40993819167908535, + "grad_norm": 0.0037914991844445467, + "learning_rate": 0.26860263466651485, + "loss": 0.1098, + "num_input_tokens_seen": 10888832, + "step": 8390 + }, + { + "epoch": 0.4101824933427797, + "grad_norm": 0.003106464399024844, + "learning_rate": 0.26856656254143674, + "loss": 0.0932, + "num_input_tokens_seen": 10895328, + "step": 8395 + }, + { + "epoch": 0.410426795006474, + "grad_norm": 0.002832240192219615, + "learning_rate": 0.2685304721319047, + "loss": 0.1137, + "num_input_tokens_seen": 10901760, + "step": 8400 + }, + { + "epoch": 0.410426795006474, + "eval_loss": 0.1417899876832962, + "eval_runtime": 401.911, + "eval_samples_per_second": 90.53, + "eval_steps_per_second": 22.634, + "num_input_tokens_seen": 10901760, + "step": 8400 + }, + { + "epoch": 0.4106710966701683, + "grad_norm": 0.0018740847008302808, + "learning_rate": 0.2684943634434843, + "loss": 0.1056, + "num_input_tokens_seen": 10908224, + "step": 8405 + }, + { + "epoch": 0.41091539833386265, + "grad_norm": 0.001579017611220479, + "learning_rate": 0.268458236481744, + "loss": 0.1176, + "num_input_tokens_seen": 10914976, + "step": 8410 + }, + { + "epoch": 0.41115969999755697, + "grad_norm": 0.002169471699744463, + "learning_rate": 0.2684220912522549, + "loss": 0.134, + "num_input_tokens_seen": 10920896, + "step": 8415 + }, + { + "epoch": 0.4114040016612513, + "grad_norm": 0.0012824153527617455, + "learning_rate": 0.2683859277605913, + "loss": 0.1316, + "num_input_tokens_seen": 10926976, + "step": 8420 + }, + { + "epoch": 0.4116483033249456, + "grad_norm": 0.002342204563319683, + "learning_rate": 0.2683497460123298, + "loss": 0.1227, + "num_input_tokens_seen": 10933568, + "step": 8425 + }, + { + "epoch": 0.41189260498864, + "grad_norm": 0.0022555477917194366, + "learning_rate": 0.26831354601305013, + "loss": 0.1399, + "num_input_tokens_seen": 10939808, + "step": 8430 + }, + { + "epoch": 0.4121369066523343, + "grad_norm": 0.003004958853125572, + "learning_rate": 0.26827732776833496, + "loss": 0.1176, + "num_input_tokens_seen": 10946464, + "step": 8435 + }, + { + "epoch": 0.41238120831602865, + "grad_norm": 0.002412147121503949, + "learning_rate": 0.26824109128376944, + "loss": 0.1214, + "num_input_tokens_seen": 10952736, + "step": 8440 + }, + { + "epoch": 0.41262550997972297, + "grad_norm": 0.001923397765494883, + "learning_rate": 0.2682048365649417, + "loss": 0.1233, + "num_input_tokens_seen": 10958848, + "step": 8445 + }, + { + "epoch": 0.4128698116434173, + "grad_norm": 0.001997373066842556, + "learning_rate": 0.2681685636174428, + "loss": 0.1206, + "num_input_tokens_seen": 10965568, + "step": 8450 + }, + { + "epoch": 0.4131141133071116, + "grad_norm": 0.0018087903736159205, + "learning_rate": 0.2681322724468663, + "loss": 0.1095, + "num_input_tokens_seen": 10972160, + "step": 8455 + }, + { + "epoch": 0.41335841497080594, + "grad_norm": 0.0021285961847752333, + "learning_rate": 0.2680959630588089, + "loss": 0.0958, + "num_input_tokens_seen": 10978720, + "step": 8460 + }, + { + "epoch": 0.41360271663450027, + "grad_norm": 0.0012703731190413237, + "learning_rate": 0.26805963545886985, + "loss": 0.1018, + "num_input_tokens_seen": 10985440, + "step": 8465 + }, + { + "epoch": 0.4138470182981946, + "grad_norm": 0.003307791892439127, + "learning_rate": 0.26802328965265143, + "loss": 0.1236, + "num_input_tokens_seen": 10992320, + "step": 8470 + }, + { + "epoch": 0.4140913199618889, + "grad_norm": 0.0038223962765187025, + "learning_rate": 0.26798692564575854, + "loss": 0.1274, + "num_input_tokens_seen": 10998432, + "step": 8475 + }, + { + "epoch": 0.4143356216255833, + "grad_norm": 0.0018129568779841065, + "learning_rate": 0.26795054344379904, + "loss": 0.1281, + "num_input_tokens_seen": 11004896, + "step": 8480 + }, + { + "epoch": 0.4145799232892776, + "grad_norm": 0.001770673436112702, + "learning_rate": 0.2679141430523835, + "loss": 0.1227, + "num_input_tokens_seen": 11011168, + "step": 8485 + }, + { + "epoch": 0.41482422495297194, + "grad_norm": 0.00415611919015646, + "learning_rate": 0.2678777244771252, + "loss": 0.1458, + "num_input_tokens_seen": 11017472, + "step": 8490 + }, + { + "epoch": 0.41506852661666627, + "grad_norm": 0.0018737594364210963, + "learning_rate": 0.2678412877236405, + "loss": 0.11, + "num_input_tokens_seen": 11023968, + "step": 8495 + }, + { + "epoch": 0.4153128282803606, + "grad_norm": 0.0024082374293357134, + "learning_rate": 0.2678048327975484, + "loss": 0.1271, + "num_input_tokens_seen": 11030784, + "step": 8500 + }, + { + "epoch": 0.4155571299440549, + "grad_norm": 0.0015394017100334167, + "learning_rate": 0.2677683597044706, + "loss": 0.1246, + "num_input_tokens_seen": 11037472, + "step": 8505 + }, + { + "epoch": 0.41580143160774924, + "grad_norm": 0.0017078814562410116, + "learning_rate": 0.2677318684500318, + "loss": 0.1011, + "num_input_tokens_seen": 11043616, + "step": 8510 + }, + { + "epoch": 0.41604573327144356, + "grad_norm": 0.0030322589445859194, + "learning_rate": 0.2676953590398593, + "loss": 0.114, + "num_input_tokens_seen": 11049984, + "step": 8515 + }, + { + "epoch": 0.4162900349351379, + "grad_norm": 0.0016455788863822818, + "learning_rate": 0.2676588314795834, + "loss": 0.1159, + "num_input_tokens_seen": 11056768, + "step": 8520 + }, + { + "epoch": 0.4165343365988322, + "grad_norm": 0.0037527307868003845, + "learning_rate": 0.26762228577483715, + "loss": 0.1221, + "num_input_tokens_seen": 11062912, + "step": 8525 + }, + { + "epoch": 0.4167786382625266, + "grad_norm": 0.0020997398532927036, + "learning_rate": 0.2675857219312563, + "loss": 0.082, + "num_input_tokens_seen": 11069856, + "step": 8530 + }, + { + "epoch": 0.4170229399262209, + "grad_norm": 0.0018376193474978209, + "learning_rate": 0.2675491399544794, + "loss": 0.0905, + "num_input_tokens_seen": 11076288, + "step": 8535 + }, + { + "epoch": 0.41726724158991524, + "grad_norm": 0.0022036924492567778, + "learning_rate": 0.2675125398501479, + "loss": 0.0945, + "num_input_tokens_seen": 11082624, + "step": 8540 + }, + { + "epoch": 0.41751154325360956, + "grad_norm": 0.0025160883087664843, + "learning_rate": 0.26747592162390604, + "loss": 0.1038, + "num_input_tokens_seen": 11088736, + "step": 8545 + }, + { + "epoch": 0.4177558449173039, + "grad_norm": 0.0023649109061807394, + "learning_rate": 0.26743928528140076, + "loss": 0.0973, + "num_input_tokens_seen": 11095360, + "step": 8550 + }, + { + "epoch": 0.4180001465809982, + "grad_norm": 0.0031786379404366016, + "learning_rate": 0.26740263082828186, + "loss": 0.1456, + "num_input_tokens_seen": 11101600, + "step": 8555 + }, + { + "epoch": 0.41824444824469253, + "grad_norm": 0.0016661053523421288, + "learning_rate": 0.2673659582702019, + "loss": 0.1235, + "num_input_tokens_seen": 11107712, + "step": 8560 + }, + { + "epoch": 0.41848874990838686, + "grad_norm": 0.0031932962592691183, + "learning_rate": 0.2673292676128163, + "loss": 0.111, + "num_input_tokens_seen": 11114144, + "step": 8565 + }, + { + "epoch": 0.4187330515720812, + "grad_norm": 0.0012970675015822053, + "learning_rate": 0.2672925588617831, + "loss": 0.1191, + "num_input_tokens_seen": 11121120, + "step": 8570 + }, + { + "epoch": 0.41897735323577556, + "grad_norm": 0.0026346698869019747, + "learning_rate": 0.2672558320227634, + "loss": 0.1165, + "num_input_tokens_seen": 11127680, + "step": 8575 + }, + { + "epoch": 0.4192216548994699, + "grad_norm": 0.0024427929893136024, + "learning_rate": 0.2672190871014209, + "loss": 0.1198, + "num_input_tokens_seen": 11134144, + "step": 8580 + }, + { + "epoch": 0.4194659565631642, + "grad_norm": 0.002220229711383581, + "learning_rate": 0.267182324103422, + "loss": 0.1112, + "num_input_tokens_seen": 11140864, + "step": 8585 + }, + { + "epoch": 0.41971025822685853, + "grad_norm": 0.002222313079982996, + "learning_rate": 0.2671455430344362, + "loss": 0.1288, + "num_input_tokens_seen": 11147264, + "step": 8590 + }, + { + "epoch": 0.41995455989055286, + "grad_norm": 0.0017383054364472628, + "learning_rate": 0.2671087439001355, + "loss": 0.1293, + "num_input_tokens_seen": 11153472, + "step": 8595 + }, + { + "epoch": 0.4201988615542472, + "grad_norm": 0.0016718142433091998, + "learning_rate": 0.2670719267061948, + "loss": 0.1264, + "num_input_tokens_seen": 11159584, + "step": 8600 + }, + { + "epoch": 0.4201988615542472, + "eval_loss": 0.12285719066858292, + "eval_runtime": 402.1741, + "eval_samples_per_second": 90.471, + "eval_steps_per_second": 22.62, + "num_input_tokens_seen": 11159584, + "step": 8600 + }, + { + "epoch": 0.4204431632179415, + "grad_norm": 0.0012926559429615736, + "learning_rate": 0.2670350914582918, + "loss": 0.1314, + "num_input_tokens_seen": 11166080, + "step": 8605 + }, + { + "epoch": 0.42068746488163583, + "grad_norm": 0.00288716540671885, + "learning_rate": 0.26699823816210694, + "loss": 0.1098, + "num_input_tokens_seen": 11172256, + "step": 8610 + }, + { + "epoch": 0.42093176654533015, + "grad_norm": 0.0028995031025260687, + "learning_rate": 0.26696136682332344, + "loss": 0.1306, + "num_input_tokens_seen": 11179136, + "step": 8615 + }, + { + "epoch": 0.4211760682090245, + "grad_norm": 0.002667696215212345, + "learning_rate": 0.2669244774476274, + "loss": 0.1075, + "num_input_tokens_seen": 11185568, + "step": 8620 + }, + { + "epoch": 0.42142036987271886, + "grad_norm": 0.0014246124774217606, + "learning_rate": 0.2668875700407075, + "loss": 0.1296, + "num_input_tokens_seen": 11192160, + "step": 8625 + }, + { + "epoch": 0.4216646715364132, + "grad_norm": 0.0019492862047627568, + "learning_rate": 0.26685064460825547, + "loss": 0.1504, + "num_input_tokens_seen": 11198336, + "step": 8630 + }, + { + "epoch": 0.4219089732001075, + "grad_norm": 0.001849992899224162, + "learning_rate": 0.26681370115596553, + "loss": 0.107, + "num_input_tokens_seen": 11204416, + "step": 8635 + }, + { + "epoch": 0.42215327486380183, + "grad_norm": 0.0013617109507322311, + "learning_rate": 0.26677673968953497, + "loss": 0.1096, + "num_input_tokens_seen": 11210688, + "step": 8640 + }, + { + "epoch": 0.42239757652749615, + "grad_norm": 0.0017919200472533703, + "learning_rate": 0.2667397602146636, + "loss": 0.1359, + "num_input_tokens_seen": 11217248, + "step": 8645 + }, + { + "epoch": 0.4226418781911905, + "grad_norm": 0.0014593367232009768, + "learning_rate": 0.2667027627370542, + "loss": 0.1096, + "num_input_tokens_seen": 11223648, + "step": 8650 + }, + { + "epoch": 0.4228861798548848, + "grad_norm": 0.001491276896558702, + "learning_rate": 0.26666574726241216, + "loss": 0.1081, + "num_input_tokens_seen": 11230208, + "step": 8655 + }, + { + "epoch": 0.4231304815185791, + "grad_norm": 0.0016823392361402512, + "learning_rate": 0.2666287137964458, + "loss": 0.136, + "num_input_tokens_seen": 11236576, + "step": 8660 + }, + { + "epoch": 0.42337478318227345, + "grad_norm": 0.0019483222858980298, + "learning_rate": 0.26659166234486614, + "loss": 0.1113, + "num_input_tokens_seen": 11242816, + "step": 8665 + }, + { + "epoch": 0.4236190848459678, + "grad_norm": 0.0022160138469189405, + "learning_rate": 0.2665545929133869, + "loss": 0.1095, + "num_input_tokens_seen": 11249504, + "step": 8670 + }, + { + "epoch": 0.42386338650966215, + "grad_norm": 0.0012175405863672495, + "learning_rate": 0.2665175055077248, + "loss": 0.1005, + "num_input_tokens_seen": 11256128, + "step": 8675 + }, + { + "epoch": 0.4241076881733565, + "grad_norm": 0.00166202278342098, + "learning_rate": 0.2664804001335991, + "loss": 0.1471, + "num_input_tokens_seen": 11263200, + "step": 8680 + }, + { + "epoch": 0.4243519898370508, + "grad_norm": 0.0015605940716341138, + "learning_rate": 0.26644327679673185, + "loss": 0.1012, + "num_input_tokens_seen": 11270048, + "step": 8685 + }, + { + "epoch": 0.4245962915007451, + "grad_norm": 0.0018200743943452835, + "learning_rate": 0.26640613550284803, + "loss": 0.1021, + "num_input_tokens_seen": 11276992, + "step": 8690 + }, + { + "epoch": 0.42484059316443945, + "grad_norm": 0.002948121167719364, + "learning_rate": 0.26636897625767525, + "loss": 0.1163, + "num_input_tokens_seen": 11283872, + "step": 8695 + }, + { + "epoch": 0.4250848948281338, + "grad_norm": 0.0043955338187515736, + "learning_rate": 0.266331799066944, + "loss": 0.1432, + "num_input_tokens_seen": 11289824, + "step": 8700 + }, + { + "epoch": 0.4253291964918281, + "grad_norm": 0.0021259335335344076, + "learning_rate": 0.2662946039363874, + "loss": 0.1233, + "num_input_tokens_seen": 11296064, + "step": 8705 + }, + { + "epoch": 0.4255734981555224, + "grad_norm": 0.0017243740148842335, + "learning_rate": 0.2662573908717414, + "loss": 0.124, + "num_input_tokens_seen": 11302368, + "step": 8710 + }, + { + "epoch": 0.42581779981921675, + "grad_norm": 0.0012987854424864054, + "learning_rate": 0.2662201598787447, + "loss": 0.114, + "num_input_tokens_seen": 11308704, + "step": 8715 + }, + { + "epoch": 0.42606210148291107, + "grad_norm": 0.002564179478213191, + "learning_rate": 0.2661829109631389, + "loss": 0.1113, + "num_input_tokens_seen": 11315360, + "step": 8720 + }, + { + "epoch": 0.42630640314660545, + "grad_norm": 0.0050730109214782715, + "learning_rate": 0.26614564413066816, + "loss": 0.1428, + "num_input_tokens_seen": 11321792, + "step": 8725 + }, + { + "epoch": 0.4265507048102998, + "grad_norm": 0.00138062983751297, + "learning_rate": 0.2661083593870795, + "loss": 0.1129, + "num_input_tokens_seen": 11328000, + "step": 8730 + }, + { + "epoch": 0.4267950064739941, + "grad_norm": 0.0014999915147200227, + "learning_rate": 0.26607105673812276, + "loss": 0.1011, + "num_input_tokens_seen": 11334560, + "step": 8735 + }, + { + "epoch": 0.4270393081376884, + "grad_norm": 0.0035878163762390614, + "learning_rate": 0.2660337361895504, + "loss": 0.1291, + "num_input_tokens_seen": 11340864, + "step": 8740 + }, + { + "epoch": 0.42728360980138275, + "grad_norm": 0.0023311516270041466, + "learning_rate": 0.26599639774711775, + "loss": 0.0812, + "num_input_tokens_seen": 11347360, + "step": 8745 + }, + { + "epoch": 0.42752791146507707, + "grad_norm": 0.002790671307593584, + "learning_rate": 0.2659590414165829, + "loss": 0.115, + "num_input_tokens_seen": 11353856, + "step": 8750 + }, + { + "epoch": 0.4277722131287714, + "grad_norm": 0.002665047300979495, + "learning_rate": 0.2659216672037066, + "loss": 0.1021, + "num_input_tokens_seen": 11360288, + "step": 8755 + }, + { + "epoch": 0.4280165147924657, + "grad_norm": 0.0015852123033255339, + "learning_rate": 0.26588427511425244, + "loss": 0.1068, + "num_input_tokens_seen": 11367040, + "step": 8760 + }, + { + "epoch": 0.42826081645616004, + "grad_norm": 0.0018599340692162514, + "learning_rate": 0.26584686515398676, + "loss": 0.119, + "num_input_tokens_seen": 11373536, + "step": 8765 + }, + { + "epoch": 0.4285051181198544, + "grad_norm": 0.001695821643806994, + "learning_rate": 0.2658094373286787, + "loss": 0.1236, + "num_input_tokens_seen": 11379808, + "step": 8770 + }, + { + "epoch": 0.42874941978354875, + "grad_norm": 0.002995660761371255, + "learning_rate": 0.2657719916441, + "loss": 0.093, + "num_input_tokens_seen": 11386784, + "step": 8775 + }, + { + "epoch": 0.42899372144724307, + "grad_norm": 0.003485628869384527, + "learning_rate": 0.2657345281060253, + "loss": 0.1584, + "num_input_tokens_seen": 11393536, + "step": 8780 + }, + { + "epoch": 0.4292380231109374, + "grad_norm": 0.0021771183237433434, + "learning_rate": 0.26569704672023203, + "loss": 0.1145, + "num_input_tokens_seen": 11399808, + "step": 8785 + }, + { + "epoch": 0.4294823247746317, + "grad_norm": 0.0013848635135218501, + "learning_rate": 0.26565954749250015, + "loss": 0.0905, + "num_input_tokens_seen": 11407648, + "step": 8790 + }, + { + "epoch": 0.42972662643832604, + "grad_norm": 0.0029214471578598022, + "learning_rate": 0.2656220304286126, + "loss": 0.1239, + "num_input_tokens_seen": 11414368, + "step": 8795 + }, + { + "epoch": 0.42997092810202037, + "grad_norm": 0.0018665335373952985, + "learning_rate": 0.265584495534355, + "loss": 0.1273, + "num_input_tokens_seen": 11420640, + "step": 8800 + }, + { + "epoch": 0.42997092810202037, + "eval_loss": 0.11826534569263458, + "eval_runtime": 402.1628, + "eval_samples_per_second": 90.473, + "eval_steps_per_second": 22.62, + "num_input_tokens_seen": 11420640, + "step": 8800 + }, + { + "epoch": 0.4302152297657147, + "grad_norm": 0.0025163046084344387, + "learning_rate": 0.2655469428155156, + "loss": 0.1028, + "num_input_tokens_seen": 11427168, + "step": 8805 + }, + { + "epoch": 0.430459531429409, + "grad_norm": 0.0011722870403900743, + "learning_rate": 0.2655093722778856, + "loss": 0.0958, + "num_input_tokens_seen": 11433856, + "step": 8810 + }, + { + "epoch": 0.43070383309310334, + "grad_norm": 0.0015859046252444386, + "learning_rate": 0.2654717839272588, + "loss": 0.1205, + "num_input_tokens_seen": 11440192, + "step": 8815 + }, + { + "epoch": 0.4309481347567977, + "grad_norm": 0.002255168044939637, + "learning_rate": 0.2654341777694318, + "loss": 0.1467, + "num_input_tokens_seen": 11446464, + "step": 8820 + }, + { + "epoch": 0.43119243642049204, + "grad_norm": 0.002421680837869644, + "learning_rate": 0.265396553810204, + "loss": 0.1275, + "num_input_tokens_seen": 11453024, + "step": 8825 + }, + { + "epoch": 0.43143673808418637, + "grad_norm": 0.0019156846683472395, + "learning_rate": 0.26535891205537737, + "loss": 0.1377, + "num_input_tokens_seen": 11459136, + "step": 8830 + }, + { + "epoch": 0.4316810397478807, + "grad_norm": 0.001887961057946086, + "learning_rate": 0.26532125251075683, + "loss": 0.1389, + "num_input_tokens_seen": 11465664, + "step": 8835 + }, + { + "epoch": 0.431925341411575, + "grad_norm": 0.0019274724181741476, + "learning_rate": 0.26528357518214996, + "loss": 0.1214, + "num_input_tokens_seen": 11472320, + "step": 8840 + }, + { + "epoch": 0.43216964307526934, + "grad_norm": 0.002570975571870804, + "learning_rate": 0.26524588007536704, + "loss": 0.1248, + "num_input_tokens_seen": 11478944, + "step": 8845 + }, + { + "epoch": 0.43241394473896366, + "grad_norm": 0.0017025596462190151, + "learning_rate": 0.26520816719622115, + "loss": 0.1205, + "num_input_tokens_seen": 11485568, + "step": 8850 + }, + { + "epoch": 0.432658246402658, + "grad_norm": 0.0017411599401384592, + "learning_rate": 0.2651704365505281, + "loss": 0.1114, + "num_input_tokens_seen": 11492448, + "step": 8855 + }, + { + "epoch": 0.4329025480663523, + "grad_norm": 0.0018653451697900891, + "learning_rate": 0.26513268814410634, + "loss": 0.1264, + "num_input_tokens_seen": 11498848, + "step": 8860 + }, + { + "epoch": 0.43314684973004663, + "grad_norm": 0.0032851763535290956, + "learning_rate": 0.2650949219827773, + "loss": 0.1309, + "num_input_tokens_seen": 11505408, + "step": 8865 + }, + { + "epoch": 0.433391151393741, + "grad_norm": 0.002683092374354601, + "learning_rate": 0.26505713807236486, + "loss": 0.1113, + "num_input_tokens_seen": 11512128, + "step": 8870 + }, + { + "epoch": 0.43363545305743534, + "grad_norm": 0.0024909055791795254, + "learning_rate": 0.26501933641869585, + "loss": 0.1103, + "num_input_tokens_seen": 11518816, + "step": 8875 + }, + { + "epoch": 0.43387975472112966, + "grad_norm": 0.001551453024148941, + "learning_rate": 0.26498151702759976, + "loss": 0.096, + "num_input_tokens_seen": 11525632, + "step": 8880 + }, + { + "epoch": 0.434124056384824, + "grad_norm": 0.0026416853070259094, + "learning_rate": 0.2649436799049088, + "loss": 0.1353, + "num_input_tokens_seen": 11531840, + "step": 8885 + }, + { + "epoch": 0.4343683580485183, + "grad_norm": 0.003487675217911601, + "learning_rate": 0.2649058250564579, + "loss": 0.1062, + "num_input_tokens_seen": 11538144, + "step": 8890 + }, + { + "epoch": 0.43461265971221263, + "grad_norm": 0.0021253693848848343, + "learning_rate": 0.26486795248808476, + "loss": 0.1364, + "num_input_tokens_seen": 11545280, + "step": 8895 + }, + { + "epoch": 0.43485696137590696, + "grad_norm": 0.0017142021097242832, + "learning_rate": 0.2648300622056298, + "loss": 0.1008, + "num_input_tokens_seen": 11552032, + "step": 8900 + }, + { + "epoch": 0.4351012630396013, + "grad_norm": 0.0009975186549127102, + "learning_rate": 0.2647921542149363, + "loss": 0.1253, + "num_input_tokens_seen": 11558464, + "step": 8905 + }, + { + "epoch": 0.4353455647032956, + "grad_norm": 0.001429248950444162, + "learning_rate": 0.26475422852185, + "loss": 0.1098, + "num_input_tokens_seen": 11565280, + "step": 8910 + }, + { + "epoch": 0.43558986636699, + "grad_norm": 0.0013571723829954863, + "learning_rate": 0.2647162851322196, + "loss": 0.1182, + "num_input_tokens_seen": 11571680, + "step": 8915 + }, + { + "epoch": 0.4358341680306843, + "grad_norm": 0.001886855810880661, + "learning_rate": 0.2646783240518964, + "loss": 0.1006, + "num_input_tokens_seen": 11577856, + "step": 8920 + }, + { + "epoch": 0.43607846969437863, + "grad_norm": 0.0022957175970077515, + "learning_rate": 0.26464034528673447, + "loss": 0.1247, + "num_input_tokens_seen": 11584768, + "step": 8925 + }, + { + "epoch": 0.43632277135807296, + "grad_norm": 0.002662458922713995, + "learning_rate": 0.26460234884259065, + "loss": 0.0938, + "num_input_tokens_seen": 11591136, + "step": 8930 + }, + { + "epoch": 0.4365670730217673, + "grad_norm": 0.0018211626447737217, + "learning_rate": 0.2645643347253245, + "loss": 0.1026, + "num_input_tokens_seen": 11598176, + "step": 8935 + }, + { + "epoch": 0.4368113746854616, + "grad_norm": 0.0022534679155796766, + "learning_rate": 0.2645263029407982, + "loss": 0.0996, + "num_input_tokens_seen": 11604384, + "step": 8940 + }, + { + "epoch": 0.43705567634915593, + "grad_norm": 0.0027752716559916735, + "learning_rate": 0.2644882534948767, + "loss": 0.1219, + "num_input_tokens_seen": 11611104, + "step": 8945 + }, + { + "epoch": 0.43729997801285025, + "grad_norm": 0.0015400294214487076, + "learning_rate": 0.2644501863934278, + "loss": 0.1017, + "num_input_tokens_seen": 11617248, + "step": 8950 + }, + { + "epoch": 0.4375442796765446, + "grad_norm": 0.001632792060263455, + "learning_rate": 0.26441210164232193, + "loss": 0.0959, + "num_input_tokens_seen": 11623936, + "step": 8955 + }, + { + "epoch": 0.4377885813402389, + "grad_norm": 0.002623960142955184, + "learning_rate": 0.26437399924743216, + "loss": 0.113, + "num_input_tokens_seen": 11630432, + "step": 8960 + }, + { + "epoch": 0.4380328830039333, + "grad_norm": 0.0013759031426161528, + "learning_rate": 0.26433587921463436, + "loss": 0.1086, + "num_input_tokens_seen": 11636992, + "step": 8965 + }, + { + "epoch": 0.4382771846676276, + "grad_norm": 0.002036179881542921, + "learning_rate": 0.2642977415498072, + "loss": 0.0998, + "num_input_tokens_seen": 11643200, + "step": 8970 + }, + { + "epoch": 0.43852148633132193, + "grad_norm": 0.001788203022442758, + "learning_rate": 0.26425958625883195, + "loss": 0.1197, + "num_input_tokens_seen": 11650240, + "step": 8975 + }, + { + "epoch": 0.43876578799501625, + "grad_norm": 0.0017959765391424298, + "learning_rate": 0.2642214133475926, + "loss": 0.1114, + "num_input_tokens_seen": 11657216, + "step": 8980 + }, + { + "epoch": 0.4390100896587106, + "grad_norm": 0.0018187452806159854, + "learning_rate": 0.26418322282197587, + "loss": 0.0947, + "num_input_tokens_seen": 11663808, + "step": 8985 + }, + { + "epoch": 0.4392543913224049, + "grad_norm": 0.002429420128464699, + "learning_rate": 0.2641450146878714, + "loss": 0.107, + "num_input_tokens_seen": 11670048, + "step": 8990 + }, + { + "epoch": 0.4394986929860992, + "grad_norm": 0.0037240502424538136, + "learning_rate": 0.26410678895117107, + "loss": 0.1328, + "num_input_tokens_seen": 11676480, + "step": 8995 + }, + { + "epoch": 0.43974299464979355, + "grad_norm": 0.0018834053771570325, + "learning_rate": 0.26406854561777, + "loss": 0.1054, + "num_input_tokens_seen": 11683072, + "step": 9000 + }, + { + "epoch": 0.43974299464979355, + "eval_loss": 0.12034117430448532, + "eval_runtime": 402.2531, + "eval_samples_per_second": 90.453, + "eval_steps_per_second": 22.615, + "num_input_tokens_seen": 11683072, + "step": 9000 + }, + { + "epoch": 0.4399872963134879, + "grad_norm": 0.002557415049523115, + "learning_rate": 0.26403028469356576, + "loss": 0.101, + "num_input_tokens_seen": 11689920, + "step": 9005 + }, + { + "epoch": 0.4402315979771822, + "grad_norm": 0.002290467731654644, + "learning_rate": 0.2639920061844585, + "loss": 0.1056, + "num_input_tokens_seen": 11696448, + "step": 9010 + }, + { + "epoch": 0.4404758996408766, + "grad_norm": 0.0034132921136915684, + "learning_rate": 0.2639537100963515, + "loss": 0.1189, + "num_input_tokens_seen": 11703200, + "step": 9015 + }, + { + "epoch": 0.4407202013045709, + "grad_norm": 0.00524330185726285, + "learning_rate": 0.26391539643515033, + "loss": 0.1231, + "num_input_tokens_seen": 11709504, + "step": 9020 + }, + { + "epoch": 0.4409645029682652, + "grad_norm": 0.0022474760189652443, + "learning_rate": 0.26387706520676346, + "loss": 0.1237, + "num_input_tokens_seen": 11716224, + "step": 9025 + }, + { + "epoch": 0.44120880463195955, + "grad_norm": 0.002126373117789626, + "learning_rate": 0.26383871641710205, + "loss": 0.1448, + "num_input_tokens_seen": 11722496, + "step": 9030 + }, + { + "epoch": 0.4414531062956539, + "grad_norm": 0.0022601985838264227, + "learning_rate": 0.26380035007208, + "loss": 0.1345, + "num_input_tokens_seen": 11728768, + "step": 9035 + }, + { + "epoch": 0.4416974079593482, + "grad_norm": 0.0017375649185851216, + "learning_rate": 0.26376196617761394, + "loss": 0.1547, + "num_input_tokens_seen": 11735136, + "step": 9040 + }, + { + "epoch": 0.4419417096230425, + "grad_norm": 0.0015359192620962858, + "learning_rate": 0.263723564739623, + "loss": 0.1198, + "num_input_tokens_seen": 11741664, + "step": 9045 + }, + { + "epoch": 0.44218601128673685, + "grad_norm": 0.0022156445775181055, + "learning_rate": 0.2636851457640293, + "loss": 0.1125, + "num_input_tokens_seen": 11748256, + "step": 9050 + }, + { + "epoch": 0.44243031295043117, + "grad_norm": 0.0023132520727813244, + "learning_rate": 0.26364670925675737, + "loss": 0.114, + "num_input_tokens_seen": 11754752, + "step": 9055 + }, + { + "epoch": 0.4426746146141255, + "grad_norm": 0.001171464566141367, + "learning_rate": 0.2636082552237347, + "loss": 0.0934, + "num_input_tokens_seen": 11760992, + "step": 9060 + }, + { + "epoch": 0.4429189162778199, + "grad_norm": 0.0010063581867143512, + "learning_rate": 0.26356978367089146, + "loss": 0.1042, + "num_input_tokens_seen": 11767616, + "step": 9065 + }, + { + "epoch": 0.4431632179415142, + "grad_norm": 0.0011378675699234009, + "learning_rate": 0.26353129460416036, + "loss": 0.1393, + "num_input_tokens_seen": 11774368, + "step": 9070 + }, + { + "epoch": 0.4434075196052085, + "grad_norm": 0.0035764267668128014, + "learning_rate": 0.2634927880294769, + "loss": 0.1202, + "num_input_tokens_seen": 11780736, + "step": 9075 + }, + { + "epoch": 0.44365182126890285, + "grad_norm": 0.0015640412457287312, + "learning_rate": 0.26345426395277927, + "loss": 0.1267, + "num_input_tokens_seen": 11787040, + "step": 9080 + }, + { + "epoch": 0.44389612293259717, + "grad_norm": 0.0012411418138071895, + "learning_rate": 0.2634157223800084, + "loss": 0.1146, + "num_input_tokens_seen": 11793376, + "step": 9085 + }, + { + "epoch": 0.4441404245962915, + "grad_norm": 0.0010976764606311917, + "learning_rate": 0.26337716331710787, + "loss": 0.1213, + "num_input_tokens_seen": 11799904, + "step": 9090 + }, + { + "epoch": 0.4443847262599858, + "grad_norm": 0.0014623973984271288, + "learning_rate": 0.2633385867700239, + "loss": 0.087, + "num_input_tokens_seen": 11806368, + "step": 9095 + }, + { + "epoch": 0.44462902792368014, + "grad_norm": 0.00239299307577312, + "learning_rate": 0.2632999927447056, + "loss": 0.145, + "num_input_tokens_seen": 11812928, + "step": 9100 + }, + { + "epoch": 0.44487332958737447, + "grad_norm": 0.0017962581478059292, + "learning_rate": 0.2632613812471046, + "loss": 0.1327, + "num_input_tokens_seen": 11819072, + "step": 9105 + }, + { + "epoch": 0.44511763125106885, + "grad_norm": 0.0014893646584823728, + "learning_rate": 0.2632227522831753, + "loss": 0.11, + "num_input_tokens_seen": 11825120, + "step": 9110 + }, + { + "epoch": 0.44536193291476317, + "grad_norm": 0.0019630859605968, + "learning_rate": 0.26318410585887475, + "loss": 0.1133, + "num_input_tokens_seen": 11831936, + "step": 9115 + }, + { + "epoch": 0.4456062345784575, + "grad_norm": 0.0028764407616108656, + "learning_rate": 0.2631454419801627, + "loss": 0.1139, + "num_input_tokens_seen": 11838624, + "step": 9120 + }, + { + "epoch": 0.4458505362421518, + "grad_norm": 0.0013777747517451644, + "learning_rate": 0.2631067606530016, + "loss": 0.1023, + "num_input_tokens_seen": 11844576, + "step": 9125 + }, + { + "epoch": 0.44609483790584614, + "grad_norm": 0.002590556163340807, + "learning_rate": 0.2630680618833567, + "loss": 0.1289, + "num_input_tokens_seen": 11851072, + "step": 9130 + }, + { + "epoch": 0.44633913956954047, + "grad_norm": 0.002640369813889265, + "learning_rate": 0.26302934567719566, + "loss": 0.1313, + "num_input_tokens_seen": 11856640, + "step": 9135 + }, + { + "epoch": 0.4465834412332348, + "grad_norm": 0.0022120848298072815, + "learning_rate": 0.2629906120404892, + "loss": 0.1246, + "num_input_tokens_seen": 11863392, + "step": 9140 + }, + { + "epoch": 0.4468277428969291, + "grad_norm": 0.001769366324879229, + "learning_rate": 0.26295186097921036, + "loss": 0.1097, + "num_input_tokens_seen": 11869984, + "step": 9145 + }, + { + "epoch": 0.44707204456062344, + "grad_norm": 0.0019285785965621471, + "learning_rate": 0.2629130924993351, + "loss": 0.1121, + "num_input_tokens_seen": 11877312, + "step": 9150 + }, + { + "epoch": 0.44731634622431776, + "grad_norm": 0.0013378934236243367, + "learning_rate": 0.2628743066068421, + "loss": 0.1013, + "num_input_tokens_seen": 11883904, + "step": 9155 + }, + { + "epoch": 0.44756064788801214, + "grad_norm": 0.0020684958435595036, + "learning_rate": 0.26283550330771244, + "loss": 0.0972, + "num_input_tokens_seen": 11890368, + "step": 9160 + }, + { + "epoch": 0.44780494955170647, + "grad_norm": 0.002584735630080104, + "learning_rate": 0.2627966826079303, + "loss": 0.1373, + "num_input_tokens_seen": 11896640, + "step": 9165 + }, + { + "epoch": 0.4480492512154008, + "grad_norm": 0.0023232433013617992, + "learning_rate": 0.26275784451348216, + "loss": 0.1128, + "num_input_tokens_seen": 11902752, + "step": 9170 + }, + { + "epoch": 0.4482935528790951, + "grad_norm": 0.0010321438312530518, + "learning_rate": 0.2627189890303574, + "loss": 0.0858, + "num_input_tokens_seen": 11909376, + "step": 9175 + }, + { + "epoch": 0.44853785454278944, + "grad_norm": 0.0019698869436979294, + "learning_rate": 0.262680116164548, + "loss": 0.1225, + "num_input_tokens_seen": 11915680, + "step": 9180 + }, + { + "epoch": 0.44878215620648376, + "grad_norm": 0.001826184568926692, + "learning_rate": 0.2626412259220487, + "loss": 0.1151, + "num_input_tokens_seen": 11922048, + "step": 9185 + }, + { + "epoch": 0.4490264578701781, + "grad_norm": 0.0016570942243561149, + "learning_rate": 0.2626023183088568, + "loss": 0.1261, + "num_input_tokens_seen": 11928448, + "step": 9190 + }, + { + "epoch": 0.4492707595338724, + "grad_norm": 0.0014291489496827126, + "learning_rate": 0.26256339333097234, + "loss": 0.114, + "num_input_tokens_seen": 11934816, + "step": 9195 + }, + { + "epoch": 0.44951506119756673, + "grad_norm": 0.0011657498544082046, + "learning_rate": 0.2625244509943981, + "loss": 0.0973, + "num_input_tokens_seen": 11941600, + "step": 9200 + }, + { + "epoch": 0.44951506119756673, + "eval_loss": 0.1146330013871193, + "eval_runtime": 402.1407, + "eval_samples_per_second": 90.478, + "eval_steps_per_second": 22.621, + "num_input_tokens_seen": 11941600, + "step": 9200 + }, + { + "epoch": 0.44975936286126106, + "grad_norm": 0.0011073033092543483, + "learning_rate": 0.2624854913051395, + "loss": 0.1189, + "num_input_tokens_seen": 11948064, + "step": 9205 + }, + { + "epoch": 0.45000366452495544, + "grad_norm": 0.0017226141644641757, + "learning_rate": 0.26244651426920446, + "loss": 0.1168, + "num_input_tokens_seen": 11954624, + "step": 9210 + }, + { + "epoch": 0.45024796618864976, + "grad_norm": 0.0023958920501172543, + "learning_rate": 0.26240751989260386, + "loss": 0.1051, + "num_input_tokens_seen": 11961184, + "step": 9215 + }, + { + "epoch": 0.4504922678523441, + "grad_norm": 0.0036157863214612007, + "learning_rate": 0.2623685081813511, + "loss": 0.1245, + "num_input_tokens_seen": 11967136, + "step": 9220 + }, + { + "epoch": 0.4507365695160384, + "grad_norm": 0.003134384984150529, + "learning_rate": 0.2623294791414623, + "loss": 0.1145, + "num_input_tokens_seen": 11973408, + "step": 9225 + }, + { + "epoch": 0.45098087117973273, + "grad_norm": 0.0010734651004895568, + "learning_rate": 0.26229043277895614, + "loss": 0.084, + "num_input_tokens_seen": 11979840, + "step": 9230 + }, + { + "epoch": 0.45122517284342706, + "grad_norm": 0.0016958445776253939, + "learning_rate": 0.2622513690998542, + "loss": 0.0897, + "num_input_tokens_seen": 11986272, + "step": 9235 + }, + { + "epoch": 0.4514694745071214, + "grad_norm": 0.00256561697460711, + "learning_rate": 0.26221228811018044, + "loss": 0.1069, + "num_input_tokens_seen": 11992928, + "step": 9240 + }, + { + "epoch": 0.4517137761708157, + "grad_norm": 0.002256878651678562, + "learning_rate": 0.2621731898159617, + "loss": 0.1107, + "num_input_tokens_seen": 11999584, + "step": 9245 + }, + { + "epoch": 0.45195807783451003, + "grad_norm": 0.001957976259291172, + "learning_rate": 0.26213407422322743, + "loss": 0.1091, + "num_input_tokens_seen": 12005984, + "step": 9250 + }, + { + "epoch": 0.4522023794982044, + "grad_norm": 0.002139823976904154, + "learning_rate": 0.2620949413380098, + "loss": 0.1345, + "num_input_tokens_seen": 12012288, + "step": 9255 + }, + { + "epoch": 0.45244668116189873, + "grad_norm": 0.0032425750978291035, + "learning_rate": 0.26205579116634353, + "loss": 0.1329, + "num_input_tokens_seen": 12018560, + "step": 9260 + }, + { + "epoch": 0.45269098282559306, + "grad_norm": 0.0021802254486829042, + "learning_rate": 0.26201662371426604, + "loss": 0.121, + "num_input_tokens_seen": 12024864, + "step": 9265 + }, + { + "epoch": 0.4529352844892874, + "grad_norm": 0.0011156175751239061, + "learning_rate": 0.2619774389878175, + "loss": 0.1244, + "num_input_tokens_seen": 12031520, + "step": 9270 + }, + { + "epoch": 0.4531795861529817, + "grad_norm": 0.001094771665520966, + "learning_rate": 0.2619382369930407, + "loss": 0.0871, + "num_input_tokens_seen": 12038048, + "step": 9275 + }, + { + "epoch": 0.45342388781667603, + "grad_norm": 0.0014694473939016461, + "learning_rate": 0.261899017735981, + "loss": 0.0889, + "num_input_tokens_seen": 12044512, + "step": 9280 + }, + { + "epoch": 0.45366818948037035, + "grad_norm": 0.001670975354500115, + "learning_rate": 0.2618597812226866, + "loss": 0.1059, + "num_input_tokens_seen": 12051072, + "step": 9285 + }, + { + "epoch": 0.4539124911440647, + "grad_norm": 0.0028907645028084517, + "learning_rate": 0.2618205274592082, + "loss": 0.1264, + "num_input_tokens_seen": 12057664, + "step": 9290 + }, + { + "epoch": 0.454156792807759, + "grad_norm": 0.0020426781848073006, + "learning_rate": 0.2617812564515992, + "loss": 0.1131, + "num_input_tokens_seen": 12063776, + "step": 9295 + }, + { + "epoch": 0.4544010944714533, + "grad_norm": 0.002740218536928296, + "learning_rate": 0.2617419682059158, + "loss": 0.121, + "num_input_tokens_seen": 12070080, + "step": 9300 + }, + { + "epoch": 0.4546453961351477, + "grad_norm": 0.001886610989458859, + "learning_rate": 0.26170266272821663, + "loss": 0.1198, + "num_input_tokens_seen": 12076192, + "step": 9305 + }, + { + "epoch": 0.45488969779884203, + "grad_norm": 0.002268022857606411, + "learning_rate": 0.26166334002456315, + "loss": 0.1101, + "num_input_tokens_seen": 12082464, + "step": 9310 + }, + { + "epoch": 0.45513399946253635, + "grad_norm": 0.0026142869610339403, + "learning_rate": 0.2616240001010194, + "loss": 0.127, + "num_input_tokens_seen": 12088672, + "step": 9315 + }, + { + "epoch": 0.4553783011262307, + "grad_norm": 0.0013029661495238543, + "learning_rate": 0.26158464296365197, + "loss": 0.1138, + "num_input_tokens_seen": 12094816, + "step": 9320 + }, + { + "epoch": 0.455622602789925, + "grad_norm": 0.002033480443060398, + "learning_rate": 0.2615452686185304, + "loss": 0.1013, + "num_input_tokens_seen": 12101152, + "step": 9325 + }, + { + "epoch": 0.4558669044536193, + "grad_norm": 0.0014218236319720745, + "learning_rate": 0.26150587707172673, + "loss": 0.1119, + "num_input_tokens_seen": 12107712, + "step": 9330 + }, + { + "epoch": 0.45611120611731365, + "grad_norm": 0.0014370223507285118, + "learning_rate": 0.2614664683293154, + "loss": 0.128, + "num_input_tokens_seen": 12114048, + "step": 9335 + }, + { + "epoch": 0.456355507781008, + "grad_norm": 0.002167262602597475, + "learning_rate": 0.26142704239737397, + "loss": 0.1511, + "num_input_tokens_seen": 12120160, + "step": 9340 + }, + { + "epoch": 0.4565998094447023, + "grad_norm": 0.0028348069172352552, + "learning_rate": 0.26138759928198235, + "loss": 0.1016, + "num_input_tokens_seen": 12126304, + "step": 9345 + }, + { + "epoch": 0.4568441111083966, + "grad_norm": 0.0014207769418135285, + "learning_rate": 0.26134813898922304, + "loss": 0.0959, + "num_input_tokens_seen": 12132544, + "step": 9350 + }, + { + "epoch": 0.457088412772091, + "grad_norm": 0.0017334086587652564, + "learning_rate": 0.26130866152518145, + "loss": 0.1082, + "num_input_tokens_seen": 12139168, + "step": 9355 + }, + { + "epoch": 0.4573327144357853, + "grad_norm": 0.0022396938875317574, + "learning_rate": 0.2612691668959455, + "loss": 0.1077, + "num_input_tokens_seen": 12146240, + "step": 9360 + }, + { + "epoch": 0.45757701609947965, + "grad_norm": 0.0022876220755279064, + "learning_rate": 0.2612296551076057, + "loss": 0.1289, + "num_input_tokens_seen": 12152768, + "step": 9365 + }, + { + "epoch": 0.457821317763174, + "grad_norm": 0.0015683344099670649, + "learning_rate": 0.26119012616625525, + "loss": 0.0961, + "num_input_tokens_seen": 12159392, + "step": 9370 + }, + { + "epoch": 0.4580656194268683, + "grad_norm": 0.0016378553118556738, + "learning_rate": 0.26115058007799, + "loss": 0.1211, + "num_input_tokens_seen": 12165600, + "step": 9375 + }, + { + "epoch": 0.4583099210905626, + "grad_norm": 0.0021402479615062475, + "learning_rate": 0.26111101684890864, + "loss": 0.0991, + "num_input_tokens_seen": 12172320, + "step": 9380 + }, + { + "epoch": 0.45855422275425695, + "grad_norm": 0.0013484145747497678, + "learning_rate": 0.26107143648511205, + "loss": 0.1045, + "num_input_tokens_seen": 12178400, + "step": 9385 + }, + { + "epoch": 0.45879852441795127, + "grad_norm": 0.0013505170354619622, + "learning_rate": 0.2610318389927042, + "loss": 0.0954, + "num_input_tokens_seen": 12185088, + "step": 9390 + }, + { + "epoch": 0.4590428260816456, + "grad_norm": 0.001789761008694768, + "learning_rate": 0.26099222437779146, + "loss": 0.0971, + "num_input_tokens_seen": 12192032, + "step": 9395 + }, + { + "epoch": 0.4592871277453399, + "grad_norm": 0.003130918135866523, + "learning_rate": 0.26095259264648285, + "loss": 0.1241, + "num_input_tokens_seen": 12198528, + "step": 9400 + }, + { + "epoch": 0.4592871277453399, + "eval_loss": 0.11861297488212585, + "eval_runtime": 402.77, + "eval_samples_per_second": 90.337, + "eval_steps_per_second": 22.586, + "num_input_tokens_seen": 12198528, + "step": 9400 + }, + { + "epoch": 0.4595314294090343, + "grad_norm": 0.0015851401258260012, + "learning_rate": 0.2609129438048902, + "loss": 0.1066, + "num_input_tokens_seen": 12204928, + "step": 9405 + }, + { + "epoch": 0.4597757310727286, + "grad_norm": 0.0023109375033527613, + "learning_rate": 0.2608732778591278, + "loss": 0.1299, + "num_input_tokens_seen": 12211136, + "step": 9410 + }, + { + "epoch": 0.46002003273642295, + "grad_norm": 0.002475339686498046, + "learning_rate": 0.2608335948153126, + "loss": 0.124, + "num_input_tokens_seen": 12217408, + "step": 9415 + }, + { + "epoch": 0.46026433440011727, + "grad_norm": 0.0019865883514285088, + "learning_rate": 0.26079389467956426, + "loss": 0.1263, + "num_input_tokens_seen": 12223392, + "step": 9420 + }, + { + "epoch": 0.4605086360638116, + "grad_norm": 0.0010653880890458822, + "learning_rate": 0.26075417745800505, + "loss": 0.1061, + "num_input_tokens_seen": 12229824, + "step": 9425 + }, + { + "epoch": 0.4607529377275059, + "grad_norm": 0.001659730332903564, + "learning_rate": 0.26071444315675985, + "loss": 0.103, + "num_input_tokens_seen": 12236128, + "step": 9430 + }, + { + "epoch": 0.46099723939120024, + "grad_norm": 0.002870383206754923, + "learning_rate": 0.2606746917819562, + "loss": 0.0911, + "num_input_tokens_seen": 12242816, + "step": 9435 + }, + { + "epoch": 0.46124154105489457, + "grad_norm": 0.0027968939393758774, + "learning_rate": 0.2606349233397242, + "loss": 0.12, + "num_input_tokens_seen": 12249184, + "step": 9440 + }, + { + "epoch": 0.4614858427185889, + "grad_norm": 0.0018239696510136127, + "learning_rate": 0.26059513783619676, + "loss": 0.1479, + "num_input_tokens_seen": 12255456, + "step": 9445 + }, + { + "epoch": 0.46173014438228327, + "grad_norm": 0.0013693226501345634, + "learning_rate": 0.26055533527750924, + "loss": 0.1021, + "num_input_tokens_seen": 12261952, + "step": 9450 + }, + { + "epoch": 0.4619744460459776, + "grad_norm": 0.0015338868834078312, + "learning_rate": 0.26051551566979964, + "loss": 0.1153, + "num_input_tokens_seen": 12268384, + "step": 9455 + }, + { + "epoch": 0.4622187477096719, + "grad_norm": 0.00217537977732718, + "learning_rate": 0.26047567901920876, + "loss": 0.1018, + "num_input_tokens_seen": 12274720, + "step": 9460 + }, + { + "epoch": 0.46246304937336624, + "grad_norm": 0.0016157124191522598, + "learning_rate": 0.2604358253318798, + "loss": 0.1207, + "num_input_tokens_seen": 12281120, + "step": 9465 + }, + { + "epoch": 0.46270735103706057, + "grad_norm": 0.0023394664749503136, + "learning_rate": 0.26039595461395876, + "loss": 0.1243, + "num_input_tokens_seen": 12287456, + "step": 9470 + }, + { + "epoch": 0.4629516527007549, + "grad_norm": 0.0016941194189712405, + "learning_rate": 0.26035606687159424, + "loss": 0.088, + "num_input_tokens_seen": 12294496, + "step": 9475 + }, + { + "epoch": 0.4631959543644492, + "grad_norm": 0.0024506286717951298, + "learning_rate": 0.26031616211093733, + "loss": 0.1264, + "num_input_tokens_seen": 12301152, + "step": 9480 + }, + { + "epoch": 0.46344025602814354, + "grad_norm": 0.0021266897674649954, + "learning_rate": 0.26027624033814195, + "loss": 0.1413, + "num_input_tokens_seen": 12307232, + "step": 9485 + }, + { + "epoch": 0.46368455769183786, + "grad_norm": 0.002030021511018276, + "learning_rate": 0.2602363015593645, + "loss": 0.1112, + "num_input_tokens_seen": 12313568, + "step": 9490 + }, + { + "epoch": 0.4639288593555322, + "grad_norm": 0.0014898955123499036, + "learning_rate": 0.26019634578076395, + "loss": 0.1373, + "num_input_tokens_seen": 12320416, + "step": 9495 + }, + { + "epoch": 0.46417316101922657, + "grad_norm": 0.0013589262962341309, + "learning_rate": 0.26015637300850214, + "loss": 0.1035, + "num_input_tokens_seen": 12327456, + "step": 9500 + }, + { + "epoch": 0.4644174626829209, + "grad_norm": 0.0013907881220802665, + "learning_rate": 0.26011638324874325, + "loss": 0.0981, + "num_input_tokens_seen": 12334368, + "step": 9505 + }, + { + "epoch": 0.4646617643466152, + "grad_norm": 0.0019354267278686166, + "learning_rate": 0.2600763765076543, + "loss": 0.1293, + "num_input_tokens_seen": 12340736, + "step": 9510 + }, + { + "epoch": 0.46490606601030954, + "grad_norm": 0.003043456468731165, + "learning_rate": 0.2600363527914048, + "loss": 0.1262, + "num_input_tokens_seen": 12347296, + "step": 9515 + }, + { + "epoch": 0.46515036767400386, + "grad_norm": 0.0022584330290555954, + "learning_rate": 0.25999631210616686, + "loss": 0.0969, + "num_input_tokens_seen": 12353792, + "step": 9520 + }, + { + "epoch": 0.4653946693376982, + "grad_norm": 0.0013535235775634646, + "learning_rate": 0.25995625445811527, + "loss": 0.12, + "num_input_tokens_seen": 12360480, + "step": 9525 + }, + { + "epoch": 0.4656389710013925, + "grad_norm": 0.002541502006351948, + "learning_rate": 0.2599161798534275, + "loss": 0.1196, + "num_input_tokens_seen": 12366688, + "step": 9530 + }, + { + "epoch": 0.46588327266508683, + "grad_norm": 0.0031577800400555134, + "learning_rate": 0.25987608829828346, + "loss": 0.1299, + "num_input_tokens_seen": 12372736, + "step": 9535 + }, + { + "epoch": 0.46612757432878116, + "grad_norm": 0.0016064719529822469, + "learning_rate": 0.25983597979886586, + "loss": 0.1321, + "num_input_tokens_seen": 12379328, + "step": 9540 + }, + { + "epoch": 0.4663718759924755, + "grad_norm": 0.0027581716421991587, + "learning_rate": 0.2597958543613599, + "loss": 0.1178, + "num_input_tokens_seen": 12385888, + "step": 9545 + }, + { + "epoch": 0.46661617765616986, + "grad_norm": 0.0016666068695485592, + "learning_rate": 0.25975571199195335, + "loss": 0.1138, + "num_input_tokens_seen": 12392192, + "step": 9550 + }, + { + "epoch": 0.4668604793198642, + "grad_norm": 0.0017462203977629542, + "learning_rate": 0.25971555269683677, + "loss": 0.097, + "num_input_tokens_seen": 12398624, + "step": 9555 + }, + { + "epoch": 0.4671047809835585, + "grad_norm": 0.002779419533908367, + "learning_rate": 0.25967537648220324, + "loss": 0.1082, + "num_input_tokens_seen": 12404768, + "step": 9560 + }, + { + "epoch": 0.46734908264725283, + "grad_norm": 0.0016655056970193982, + "learning_rate": 0.2596351833542483, + "loss": 0.1234, + "num_input_tokens_seen": 12410880, + "step": 9565 + }, + { + "epoch": 0.46759338431094716, + "grad_norm": 0.0022677171509712934, + "learning_rate": 0.25959497331917036, + "loss": 0.1167, + "num_input_tokens_seen": 12417280, + "step": 9570 + }, + { + "epoch": 0.4678376859746415, + "grad_norm": 0.002046798123046756, + "learning_rate": 0.2595547463831703, + "loss": 0.1226, + "num_input_tokens_seen": 12423808, + "step": 9575 + }, + { + "epoch": 0.4680819876383358, + "grad_norm": 0.001124708796851337, + "learning_rate": 0.25951450255245156, + "loss": 0.1129, + "num_input_tokens_seen": 12430560, + "step": 9580 + }, + { + "epoch": 0.46832628930203013, + "grad_norm": 0.0025589230936020613, + "learning_rate": 0.2594742418332203, + "loss": 0.1241, + "num_input_tokens_seen": 12436992, + "step": 9585 + }, + { + "epoch": 0.46857059096572445, + "grad_norm": 0.0013295823009684682, + "learning_rate": 0.2594339642316852, + "loss": 0.1257, + "num_input_tokens_seen": 12443136, + "step": 9590 + }, + { + "epoch": 0.46881489262941883, + "grad_norm": 0.0026515605859458447, + "learning_rate": 0.2593936697540576, + "loss": 0.1472, + "num_input_tokens_seen": 12449472, + "step": 9595 + }, + { + "epoch": 0.46905919429311316, + "grad_norm": 0.0016618132358416915, + "learning_rate": 0.2593533584065514, + "loss": 0.1244, + "num_input_tokens_seen": 12455968, + "step": 9600 + }, + { + "epoch": 0.46905919429311316, + "eval_loss": 0.12838327884674072, + "eval_runtime": 402.2159, + "eval_samples_per_second": 90.461, + "eval_steps_per_second": 22.617, + "num_input_tokens_seen": 12455968, + "step": 9600 + }, + { + "epoch": 0.4693034959568075, + "grad_norm": 0.0013023335486650467, + "learning_rate": 0.2593130301953831, + "loss": 0.1312, + "num_input_tokens_seen": 12462304, + "step": 9605 + }, + { + "epoch": 0.4695477976205018, + "grad_norm": 0.0020031898748129606, + "learning_rate": 0.2592726851267718, + "loss": 0.1026, + "num_input_tokens_seen": 12469088, + "step": 9610 + }, + { + "epoch": 0.46979209928419613, + "grad_norm": 0.0010194906499236822, + "learning_rate": 0.2592323232069393, + "loss": 0.0959, + "num_input_tokens_seen": 12475296, + "step": 9615 + }, + { + "epoch": 0.47003640094789045, + "grad_norm": 0.002372491406276822, + "learning_rate": 0.25919194444210986, + "loss": 0.1172, + "num_input_tokens_seen": 12481824, + "step": 9620 + }, + { + "epoch": 0.4702807026115848, + "grad_norm": 0.001661317190155387, + "learning_rate": 0.2591515488385103, + "loss": 0.0997, + "num_input_tokens_seen": 12488928, + "step": 9625 + }, + { + "epoch": 0.4705250042752791, + "grad_norm": 0.0016091769794002175, + "learning_rate": 0.2591111364023704, + "loss": 0.1262, + "num_input_tokens_seen": 12495584, + "step": 9630 + }, + { + "epoch": 0.4707693059389734, + "grad_norm": 0.001319811912253499, + "learning_rate": 0.259070707139922, + "loss": 0.1032, + "num_input_tokens_seen": 12502336, + "step": 9635 + }, + { + "epoch": 0.47101360760266775, + "grad_norm": 0.000991817214526236, + "learning_rate": 0.25903026105739985, + "loss": 0.1102, + "num_input_tokens_seen": 12509120, + "step": 9640 + }, + { + "epoch": 0.47125790926636213, + "grad_norm": 0.0015194264706224203, + "learning_rate": 0.2589897981610413, + "loss": 0.1224, + "num_input_tokens_seen": 12515296, + "step": 9645 + }, + { + "epoch": 0.47150221093005645, + "grad_norm": 0.0027461450081318617, + "learning_rate": 0.2589493184570863, + "loss": 0.1401, + "num_input_tokens_seen": 12521792, + "step": 9650 + }, + { + "epoch": 0.4717465125937508, + "grad_norm": 0.0020115585066378117, + "learning_rate": 0.25890882195177717, + "loss": 0.1187, + "num_input_tokens_seen": 12528064, + "step": 9655 + }, + { + "epoch": 0.4719908142574451, + "grad_norm": 0.0018000085838139057, + "learning_rate": 0.25886830865135907, + "loss": 0.1212, + "num_input_tokens_seen": 12535328, + "step": 9660 + }, + { + "epoch": 0.4722351159211394, + "grad_norm": 0.0016057674074545503, + "learning_rate": 0.25882777856207967, + "loss": 0.1004, + "num_input_tokens_seen": 12542400, + "step": 9665 + }, + { + "epoch": 0.47247941758483375, + "grad_norm": 0.0015737981302663684, + "learning_rate": 0.2587872316901892, + "loss": 0.1146, + "num_input_tokens_seen": 12548832, + "step": 9670 + }, + { + "epoch": 0.4727237192485281, + "grad_norm": 0.0019519173074513674, + "learning_rate": 0.25874666804194046, + "loss": 0.1387, + "num_input_tokens_seen": 12555264, + "step": 9675 + }, + { + "epoch": 0.4729680209122224, + "grad_norm": 0.001297205570153892, + "learning_rate": 0.258706087623589, + "loss": 0.1005, + "num_input_tokens_seen": 12561312, + "step": 9680 + }, + { + "epoch": 0.4732123225759167, + "grad_norm": 0.0018688627751544118, + "learning_rate": 0.25866549044139264, + "loss": 0.0968, + "num_input_tokens_seen": 12567616, + "step": 9685 + }, + { + "epoch": 0.47345662423961105, + "grad_norm": 0.0042114038951694965, + "learning_rate": 0.25862487650161214, + "loss": 0.1109, + "num_input_tokens_seen": 12574048, + "step": 9690 + }, + { + "epoch": 0.4737009259033054, + "grad_norm": 0.0022727439645677805, + "learning_rate": 0.2585842458105106, + "loss": 0.1164, + "num_input_tokens_seen": 12580896, + "step": 9695 + }, + { + "epoch": 0.47394522756699975, + "grad_norm": 0.0009233637829311192, + "learning_rate": 0.2585435983743538, + "loss": 0.0923, + "num_input_tokens_seen": 12587552, + "step": 9700 + }, + { + "epoch": 0.4741895292306941, + "grad_norm": 0.0017936071380972862, + "learning_rate": 0.2585029341994101, + "loss": 0.1076, + "num_input_tokens_seen": 12593696, + "step": 9705 + }, + { + "epoch": 0.4744338308943884, + "grad_norm": 0.0033004742581397295, + "learning_rate": 0.2584622532919504, + "loss": 0.1227, + "num_input_tokens_seen": 12600576, + "step": 9710 + }, + { + "epoch": 0.4746781325580827, + "grad_norm": 0.0017023446271196008, + "learning_rate": 0.2584215556582482, + "loss": 0.1127, + "num_input_tokens_seen": 12606912, + "step": 9715 + }, + { + "epoch": 0.47492243422177705, + "grad_norm": 0.0014545820886269212, + "learning_rate": 0.25838084130457967, + "loss": 0.0854, + "num_input_tokens_seen": 12613312, + "step": 9720 + }, + { + "epoch": 0.47516673588547137, + "grad_norm": 0.0015381575794890523, + "learning_rate": 0.2583401102372234, + "loss": 0.1099, + "num_input_tokens_seen": 12620128, + "step": 9725 + }, + { + "epoch": 0.4754110375491657, + "grad_norm": 0.0014243015320971608, + "learning_rate": 0.2582993624624606, + "loss": 0.0982, + "num_input_tokens_seen": 12627040, + "step": 9730 + }, + { + "epoch": 0.47565533921286, + "grad_norm": 0.00251763011328876, + "learning_rate": 0.25825859798657513, + "loss": 0.1155, + "num_input_tokens_seen": 12634112, + "step": 9735 + }, + { + "epoch": 0.47589964087655434, + "grad_norm": 0.0018161151092499495, + "learning_rate": 0.25821781681585343, + "loss": 0.0851, + "num_input_tokens_seen": 12640864, + "step": 9740 + }, + { + "epoch": 0.4761439425402487, + "grad_norm": 0.0020431424491107464, + "learning_rate": 0.2581770189565844, + "loss": 0.0981, + "num_input_tokens_seen": 12647296, + "step": 9745 + }, + { + "epoch": 0.47638824420394305, + "grad_norm": 0.0018527675420045853, + "learning_rate": 0.25813620441505963, + "loss": 0.1012, + "num_input_tokens_seen": 12653632, + "step": 9750 + }, + { + "epoch": 0.47663254586763737, + "grad_norm": 0.0020877488423138857, + "learning_rate": 0.2580953731975732, + "loss": 0.1261, + "num_input_tokens_seen": 12659872, + "step": 9755 + }, + { + "epoch": 0.4768768475313317, + "grad_norm": 0.0019499447662383318, + "learning_rate": 0.2580545253104218, + "loss": 0.0878, + "num_input_tokens_seen": 12666272, + "step": 9760 + }, + { + "epoch": 0.477121149195026, + "grad_norm": 0.0015313151525333524, + "learning_rate": 0.2580136607599047, + "loss": 0.115, + "num_input_tokens_seen": 12672544, + "step": 9765 + }, + { + "epoch": 0.47736545085872034, + "grad_norm": 0.001127170748077333, + "learning_rate": 0.2579727795523238, + "loss": 0.1074, + "num_input_tokens_seen": 12678944, + "step": 9770 + }, + { + "epoch": 0.47760975252241467, + "grad_norm": 0.0013875799486413598, + "learning_rate": 0.25793188169398334, + "loss": 0.1109, + "num_input_tokens_seen": 12685568, + "step": 9775 + }, + { + "epoch": 0.477854054186109, + "grad_norm": 0.0025148168206214905, + "learning_rate": 0.25789096719119037, + "loss": 0.1047, + "num_input_tokens_seen": 12691648, + "step": 9780 + }, + { + "epoch": 0.4780983558498033, + "grad_norm": 0.001895428984425962, + "learning_rate": 0.2578500360502544, + "loss": 0.11, + "num_input_tokens_seen": 12698048, + "step": 9785 + }, + { + "epoch": 0.4783426575134977, + "grad_norm": 0.0025196673814207315, + "learning_rate": 0.2578090882774876, + "loss": 0.1063, + "num_input_tokens_seen": 12704256, + "step": 9790 + }, + { + "epoch": 0.478586959177192, + "grad_norm": 0.0020912790205329657, + "learning_rate": 0.25776812387920456, + "loss": 0.1333, + "num_input_tokens_seen": 12710560, + "step": 9795 + }, + { + "epoch": 0.47883126084088634, + "grad_norm": 0.0013552403543144464, + "learning_rate": 0.2577271428617225, + "loss": 0.1204, + "num_input_tokens_seen": 12716992, + "step": 9800 + }, + { + "epoch": 0.47883126084088634, + "eval_loss": 0.11165111511945724, + "eval_runtime": 402.7606, + "eval_samples_per_second": 90.339, + "eval_steps_per_second": 22.587, + "num_input_tokens_seen": 12716992, + "step": 9800 + }, + { + "epoch": 0.47907556250458067, + "grad_norm": 0.0034787915647029877, + "learning_rate": 0.25768614523136124, + "loss": 0.1279, + "num_input_tokens_seen": 12722976, + "step": 9805 + }, + { + "epoch": 0.479319864168275, + "grad_norm": 0.0019449025858193636, + "learning_rate": 0.25764513099444314, + "loss": 0.1103, + "num_input_tokens_seen": 12729088, + "step": 9810 + }, + { + "epoch": 0.4795641658319693, + "grad_norm": 0.0008894497877918184, + "learning_rate": 0.25760410015729307, + "loss": 0.091, + "num_input_tokens_seen": 12735456, + "step": 9815 + }, + { + "epoch": 0.47980846749566364, + "grad_norm": 0.0021570250391960144, + "learning_rate": 0.2575630527262385, + "loss": 0.1082, + "num_input_tokens_seen": 12742208, + "step": 9820 + }, + { + "epoch": 0.48005276915935796, + "grad_norm": 0.001462982501834631, + "learning_rate": 0.25752198870760945, + "loss": 0.1035, + "num_input_tokens_seen": 12748704, + "step": 9825 + }, + { + "epoch": 0.4802970708230523, + "grad_norm": 0.0016204760177060962, + "learning_rate": 0.2574809081077386, + "loss": 0.1192, + "num_input_tokens_seen": 12754784, + "step": 9830 + }, + { + "epoch": 0.4805413724867466, + "grad_norm": 0.0011354495072737336, + "learning_rate": 0.257439810932961, + "loss": 0.0881, + "num_input_tokens_seen": 12761024, + "step": 9835 + }, + { + "epoch": 0.480785674150441, + "grad_norm": 0.002648488152772188, + "learning_rate": 0.2573986971896144, + "loss": 0.1264, + "num_input_tokens_seen": 12767616, + "step": 9840 + }, + { + "epoch": 0.4810299758141353, + "grad_norm": 0.0018123837653547525, + "learning_rate": 0.257357566884039, + "loss": 0.1315, + "num_input_tokens_seen": 12774080, + "step": 9845 + }, + { + "epoch": 0.48127427747782964, + "grad_norm": 0.0017283849883824587, + "learning_rate": 0.25731642002257765, + "loss": 0.1079, + "num_input_tokens_seen": 12780352, + "step": 9850 + }, + { + "epoch": 0.48151857914152396, + "grad_norm": 0.002899334765970707, + "learning_rate": 0.25727525661157574, + "loss": 0.1358, + "num_input_tokens_seen": 12786816, + "step": 9855 + }, + { + "epoch": 0.4817628808052183, + "grad_norm": 0.002107046078890562, + "learning_rate": 0.2572340766573811, + "loss": 0.1319, + "num_input_tokens_seen": 12793696, + "step": 9860 + }, + { + "epoch": 0.4820071824689126, + "grad_norm": 0.0027606934309005737, + "learning_rate": 0.25719288016634434, + "loss": 0.1238, + "num_input_tokens_seen": 12799904, + "step": 9865 + }, + { + "epoch": 0.48225148413260693, + "grad_norm": 0.0023630838841199875, + "learning_rate": 0.25715166714481835, + "loss": 0.1327, + "num_input_tokens_seen": 12806848, + "step": 9870 + }, + { + "epoch": 0.48249578579630126, + "grad_norm": 0.0014539749827235937, + "learning_rate": 0.2571104375991587, + "loss": 0.1161, + "num_input_tokens_seen": 12813440, + "step": 9875 + }, + { + "epoch": 0.4827400874599956, + "grad_norm": 0.001291996450163424, + "learning_rate": 0.2570691915357236, + "loss": 0.1381, + "num_input_tokens_seen": 12819648, + "step": 9880 + }, + { + "epoch": 0.4829843891236899, + "grad_norm": 0.001057955319993198, + "learning_rate": 0.2570279289608736, + "loss": 0.1108, + "num_input_tokens_seen": 12825760, + "step": 9885 + }, + { + "epoch": 0.4832286907873843, + "grad_norm": 0.001002384815365076, + "learning_rate": 0.256986649880972, + "loss": 0.1236, + "num_input_tokens_seen": 12832768, + "step": 9890 + }, + { + "epoch": 0.4834729924510786, + "grad_norm": 0.0009498866857029498, + "learning_rate": 0.25694535430238447, + "loss": 0.1034, + "num_input_tokens_seen": 12838944, + "step": 9895 + }, + { + "epoch": 0.48371729411477293, + "grad_norm": 0.0024899851996451616, + "learning_rate": 0.25690404223147933, + "loss": 0.1441, + "num_input_tokens_seen": 12844992, + "step": 9900 + }, + { + "epoch": 0.48396159577846726, + "grad_norm": 0.0012028546771034598, + "learning_rate": 0.2568627136746275, + "loss": 0.1062, + "num_input_tokens_seen": 12851296, + "step": 9905 + }, + { + "epoch": 0.4842058974421616, + "grad_norm": 0.001034553861245513, + "learning_rate": 0.25682136863820226, + "loss": 0.1262, + "num_input_tokens_seen": 12857472, + "step": 9910 + }, + { + "epoch": 0.4844501991058559, + "grad_norm": 0.001249017077498138, + "learning_rate": 0.25678000712857957, + "loss": 0.0981, + "num_input_tokens_seen": 12863904, + "step": 9915 + }, + { + "epoch": 0.48469450076955023, + "grad_norm": 0.0017015141202136874, + "learning_rate": 0.2567386291521379, + "loss": 0.1417, + "num_input_tokens_seen": 12870016, + "step": 9920 + }, + { + "epoch": 0.48493880243324455, + "grad_norm": 0.002595057711005211, + "learning_rate": 0.2566972347152583, + "loss": 0.1079, + "num_input_tokens_seen": 12876800, + "step": 9925 + }, + { + "epoch": 0.4851831040969389, + "grad_norm": 0.0014667087234556675, + "learning_rate": 0.2566558238243242, + "loss": 0.123, + "num_input_tokens_seen": 12883072, + "step": 9930 + }, + { + "epoch": 0.48542740576063326, + "grad_norm": 0.0012832414358854294, + "learning_rate": 0.25661439648572176, + "loss": 0.1241, + "num_input_tokens_seen": 12889440, + "step": 9935 + }, + { + "epoch": 0.4856717074243276, + "grad_norm": 0.0014175500255078077, + "learning_rate": 0.25657295270583963, + "loss": 0.1099, + "num_input_tokens_seen": 12895744, + "step": 9940 + }, + { + "epoch": 0.4859160090880219, + "grad_norm": 0.0013960934011265635, + "learning_rate": 0.25653149249106894, + "loss": 0.1185, + "num_input_tokens_seen": 12902016, + "step": 9945 + }, + { + "epoch": 0.48616031075171623, + "grad_norm": 0.0013120381627231836, + "learning_rate": 0.25649001584780323, + "loss": 0.0863, + "num_input_tokens_seen": 12908576, + "step": 9950 + }, + { + "epoch": 0.48640461241541055, + "grad_norm": 0.001381564186885953, + "learning_rate": 0.2564485227824389, + "loss": 0.1164, + "num_input_tokens_seen": 12914688, + "step": 9955 + }, + { + "epoch": 0.4866489140791049, + "grad_norm": 0.0021526897326111794, + "learning_rate": 0.25640701330137466, + "loss": 0.1332, + "num_input_tokens_seen": 12921024, + "step": 9960 + }, + { + "epoch": 0.4868932157427992, + "grad_norm": 0.0024423464201390743, + "learning_rate": 0.2563654874110117, + "loss": 0.084, + "num_input_tokens_seen": 12927872, + "step": 9965 + }, + { + "epoch": 0.4871375174064935, + "grad_norm": 0.0011521651176735759, + "learning_rate": 0.256323945117754, + "loss": 0.1198, + "num_input_tokens_seen": 12934208, + "step": 9970 + }, + { + "epoch": 0.48738181907018785, + "grad_norm": 0.0019722143188118935, + "learning_rate": 0.2562823864280078, + "loss": 0.118, + "num_input_tokens_seen": 12940736, + "step": 9975 + }, + { + "epoch": 0.4876261207338822, + "grad_norm": 0.0016280540730804205, + "learning_rate": 0.25624081134818194, + "loss": 0.0934, + "num_input_tokens_seen": 12947360, + "step": 9980 + }, + { + "epoch": 0.48787042239757655, + "grad_norm": 0.002261681016534567, + "learning_rate": 0.2561992198846879, + "loss": 0.1344, + "num_input_tokens_seen": 12954176, + "step": 9985 + }, + { + "epoch": 0.4881147240612709, + "grad_norm": 0.0026697234716266394, + "learning_rate": 0.25615761204393955, + "loss": 0.1278, + "num_input_tokens_seen": 12960928, + "step": 9990 + }, + { + "epoch": 0.4883590257249652, + "grad_norm": 0.0022933045402169228, + "learning_rate": 0.2561159878323534, + "loss": 0.1231, + "num_input_tokens_seen": 12967648, + "step": 9995 + }, + { + "epoch": 0.4886033273886595, + "grad_norm": 0.0008182452875189483, + "learning_rate": 0.2560743472563483, + "loss": 0.0962, + "num_input_tokens_seen": 12974048, + "step": 10000 + }, + { + "epoch": 0.4886033273886595, + "eval_loss": 0.11820349842309952, + "eval_runtime": 401.8554, + "eval_samples_per_second": 90.543, + "eval_steps_per_second": 22.637, + "num_input_tokens_seen": 12974048, + "step": 10000 + }, + { + "epoch": 0.48884762905235385, + "grad_norm": 0.0018146659713238478, + "learning_rate": 0.25603269032234593, + "loss": 0.1223, + "num_input_tokens_seen": 12980960, + "step": 10005 + }, + { + "epoch": 0.4890919307160482, + "grad_norm": 0.002811427926644683, + "learning_rate": 0.2559910170367702, + "loss": 0.1359, + "num_input_tokens_seen": 12987456, + "step": 10010 + }, + { + "epoch": 0.4893362323797425, + "grad_norm": 0.0013212288031354547, + "learning_rate": 0.2559493274060477, + "loss": 0.1172, + "num_input_tokens_seen": 12994016, + "step": 10015 + }, + { + "epoch": 0.4895805340434368, + "grad_norm": 0.003728148527443409, + "learning_rate": 0.2559076214366074, + "loss": 0.1409, + "num_input_tokens_seen": 13000256, + "step": 10020 + }, + { + "epoch": 0.48982483570713115, + "grad_norm": 0.001283983699977398, + "learning_rate": 0.25586589913488106, + "loss": 0.1099, + "num_input_tokens_seen": 13006368, + "step": 10025 + }, + { + "epoch": 0.49006913737082547, + "grad_norm": 0.0012187574757263064, + "learning_rate": 0.2558241605073026, + "loss": 0.1239, + "num_input_tokens_seen": 13012736, + "step": 10030 + }, + { + "epoch": 0.49031343903451985, + "grad_norm": 0.0010102998930960894, + "learning_rate": 0.25578240556030873, + "loss": 0.1048, + "num_input_tokens_seen": 13019072, + "step": 10035 + }, + { + "epoch": 0.4905577406982142, + "grad_norm": 0.0020058350637555122, + "learning_rate": 0.2557406343003386, + "loss": 0.13, + "num_input_tokens_seen": 13025472, + "step": 10040 + }, + { + "epoch": 0.4908020423619085, + "grad_norm": 0.0012518564471974969, + "learning_rate": 0.25569884673383375, + "loss": 0.1055, + "num_input_tokens_seen": 13031840, + "step": 10045 + }, + { + "epoch": 0.4910463440256028, + "grad_norm": 0.0014018358197063208, + "learning_rate": 0.25565704286723856, + "loss": 0.1, + "num_input_tokens_seen": 13038272, + "step": 10050 + }, + { + "epoch": 0.49129064568929715, + "grad_norm": 0.0027738846838474274, + "learning_rate": 0.25561522270699955, + "loss": 0.1319, + "num_input_tokens_seen": 13044544, + "step": 10055 + }, + { + "epoch": 0.49153494735299147, + "grad_norm": 0.0021554380655288696, + "learning_rate": 0.25557338625956594, + "loss": 0.1531, + "num_input_tokens_seen": 13050848, + "step": 10060 + }, + { + "epoch": 0.4917792490166858, + "grad_norm": 0.0009380607516504824, + "learning_rate": 0.25553153353138947, + "loss": 0.1339, + "num_input_tokens_seen": 13057280, + "step": 10065 + }, + { + "epoch": 0.4920235506803801, + "grad_norm": 0.0009417547844350338, + "learning_rate": 0.2554896645289243, + "loss": 0.1346, + "num_input_tokens_seen": 13063744, + "step": 10070 + }, + { + "epoch": 0.49226785234407444, + "grad_norm": 0.0016230018809437752, + "learning_rate": 0.2554477792586272, + "loss": 0.1087, + "num_input_tokens_seen": 13070304, + "step": 10075 + }, + { + "epoch": 0.49251215400776877, + "grad_norm": 0.0009458018466830254, + "learning_rate": 0.25540587772695744, + "loss": 0.1186, + "num_input_tokens_seen": 13076416, + "step": 10080 + }, + { + "epoch": 0.49275645567146314, + "grad_norm": 0.0011661236640065908, + "learning_rate": 0.2553639599403767, + "loss": 0.1229, + "num_input_tokens_seen": 13082816, + "step": 10085 + }, + { + "epoch": 0.49300075733515747, + "grad_norm": 0.0008761748904362321, + "learning_rate": 0.2553220259053493, + "loss": 0.1005, + "num_input_tokens_seen": 13089280, + "step": 10090 + }, + { + "epoch": 0.4932450589988518, + "grad_norm": 0.002006860449910164, + "learning_rate": 0.2552800756283419, + "loss": 0.1004, + "num_input_tokens_seen": 13095680, + "step": 10095 + }, + { + "epoch": 0.4934893606625461, + "grad_norm": 0.0011931119952350855, + "learning_rate": 0.25523810911582373, + "loss": 0.1005, + "num_input_tokens_seen": 13102112, + "step": 10100 + }, + { + "epoch": 0.49373366232624044, + "grad_norm": 0.0017533862264826894, + "learning_rate": 0.25519612637426675, + "loss": 0.1258, + "num_input_tokens_seen": 13108640, + "step": 10105 + }, + { + "epoch": 0.49397796398993477, + "grad_norm": 0.0027104017790406942, + "learning_rate": 0.25515412741014504, + "loss": 0.1249, + "num_input_tokens_seen": 13115136, + "step": 10110 + }, + { + "epoch": 0.4942222656536291, + "grad_norm": 0.0018405918963253498, + "learning_rate": 0.2551121122299355, + "loss": 0.1055, + "num_input_tokens_seen": 13121600, + "step": 10115 + }, + { + "epoch": 0.4944665673173234, + "grad_norm": 0.0017820430221036077, + "learning_rate": 0.2550700808401173, + "loss": 0.107, + "num_input_tokens_seen": 13128288, + "step": 10120 + }, + { + "epoch": 0.49471086898101774, + "grad_norm": 0.002238648710772395, + "learning_rate": 0.2550280332471722, + "loss": 0.1199, + "num_input_tokens_seen": 13134272, + "step": 10125 + }, + { + "epoch": 0.4949551706447121, + "grad_norm": 0.0009641119395382702, + "learning_rate": 0.2549859694575845, + "loss": 0.1118, + "num_input_tokens_seen": 13140576, + "step": 10130 + }, + { + "epoch": 0.49519947230840644, + "grad_norm": 0.0019614319317042828, + "learning_rate": 0.254943889477841, + "loss": 0.1131, + "num_input_tokens_seen": 13147040, + "step": 10135 + }, + { + "epoch": 0.49544377397210076, + "grad_norm": 0.0024499250575900078, + "learning_rate": 0.25490179331443097, + "loss": 0.0972, + "num_input_tokens_seen": 13154016, + "step": 10140 + }, + { + "epoch": 0.4956880756357951, + "grad_norm": 0.0015940091107040644, + "learning_rate": 0.25485968097384615, + "loss": 0.0951, + "num_input_tokens_seen": 13160288, + "step": 10145 + }, + { + "epoch": 0.4959323772994894, + "grad_norm": 0.0018942024325951934, + "learning_rate": 0.25481755246258075, + "loss": 0.0996, + "num_input_tokens_seen": 13166752, + "step": 10150 + }, + { + "epoch": 0.49617667896318374, + "grad_norm": 0.0018232122529298067, + "learning_rate": 0.2547754077871315, + "loss": 0.1091, + "num_input_tokens_seen": 13173632, + "step": 10155 + }, + { + "epoch": 0.49642098062687806, + "grad_norm": 0.0018774359486997128, + "learning_rate": 0.25473324695399774, + "loss": 0.1126, + "num_input_tokens_seen": 13180512, + "step": 10160 + }, + { + "epoch": 0.4966652822905724, + "grad_norm": 0.002090892056003213, + "learning_rate": 0.25469106996968105, + "loss": 0.1067, + "num_input_tokens_seen": 13186752, + "step": 10165 + }, + { + "epoch": 0.4969095839542667, + "grad_norm": 0.001595717971213162, + "learning_rate": 0.2546488768406858, + "loss": 0.0968, + "num_input_tokens_seen": 13193152, + "step": 10170 + }, + { + "epoch": 0.49715388561796103, + "grad_norm": 0.0020495248027145863, + "learning_rate": 0.25460666757351863, + "loss": 0.1165, + "num_input_tokens_seen": 13199456, + "step": 10175 + }, + { + "epoch": 0.4973981872816554, + "grad_norm": 0.0015856140526011586, + "learning_rate": 0.25456444217468877, + "loss": 0.1116, + "num_input_tokens_seen": 13205728, + "step": 10180 + }, + { + "epoch": 0.49764248894534974, + "grad_norm": 0.0011678090086206794, + "learning_rate": 0.25452220065070785, + "loss": 0.1028, + "num_input_tokens_seen": 13212000, + "step": 10185 + }, + { + "epoch": 0.49788679060904406, + "grad_norm": 0.0028825311455875635, + "learning_rate": 0.2544799430080901, + "loss": 0.1197, + "num_input_tokens_seen": 13218048, + "step": 10190 + }, + { + "epoch": 0.4981310922727384, + "grad_norm": 0.001694979378953576, + "learning_rate": 0.2544376692533522, + "loss": 0.1158, + "num_input_tokens_seen": 13224640, + "step": 10195 + }, + { + "epoch": 0.4983753939364327, + "grad_norm": 0.0013776409905403852, + "learning_rate": 0.2543953793930132, + "loss": 0.0938, + "num_input_tokens_seen": 13231360, + "step": 10200 + }, + { + "epoch": 0.4983753939364327, + "eval_loss": 0.11140511929988861, + "eval_runtime": 402.3341, + "eval_samples_per_second": 90.435, + "eval_steps_per_second": 22.611, + "num_input_tokens_seen": 13231360, + "step": 10200 + }, + { + "epoch": 0.49861969560012703, + "grad_norm": 0.0010741668520495296, + "learning_rate": 0.2543530734335948, + "loss": 0.1177, + "num_input_tokens_seen": 13237600, + "step": 10205 + }, + { + "epoch": 0.49886399726382136, + "grad_norm": 0.001955880783498287, + "learning_rate": 0.2543107513816211, + "loss": 0.1214, + "num_input_tokens_seen": 13244256, + "step": 10210 + }, + { + "epoch": 0.4991082989275157, + "grad_norm": 0.001390855060890317, + "learning_rate": 0.25426841324361865, + "loss": 0.0758, + "num_input_tokens_seen": 13251136, + "step": 10215 + }, + { + "epoch": 0.49935260059121, + "grad_norm": 0.0018839439144358039, + "learning_rate": 0.2542260590261166, + "loss": 0.103, + "num_input_tokens_seen": 13257536, + "step": 10220 + }, + { + "epoch": 0.49959690225490433, + "grad_norm": 0.0014259060844779015, + "learning_rate": 0.2541836887356465, + "loss": 0.0851, + "num_input_tokens_seen": 13264000, + "step": 10225 + }, + { + "epoch": 0.4998412039185987, + "grad_norm": 0.0021618162281811237, + "learning_rate": 0.2541413023787423, + "loss": 0.1085, + "num_input_tokens_seen": 13270560, + "step": 10230 + }, + { + "epoch": 0.500085505582293, + "grad_norm": 0.0016483257059007883, + "learning_rate": 0.2540988999619405, + "loss": 0.0785, + "num_input_tokens_seen": 13276864, + "step": 10235 + }, + { + "epoch": 0.5003298072459873, + "grad_norm": 0.0023259338922798634, + "learning_rate": 0.25405648149178023, + "loss": 0.1397, + "num_input_tokens_seen": 13283360, + "step": 10240 + }, + { + "epoch": 0.5005741089096817, + "grad_norm": 0.0013770764926448464, + "learning_rate": 0.2540140469748028, + "loss": 0.1184, + "num_input_tokens_seen": 13290048, + "step": 10245 + }, + { + "epoch": 0.500818410573376, + "grad_norm": 0.0022583291865885258, + "learning_rate": 0.25397159641755224, + "loss": 0.1288, + "num_input_tokens_seen": 13296032, + "step": 10250 + }, + { + "epoch": 0.5010627122370703, + "grad_norm": 0.0015590139664709568, + "learning_rate": 0.2539291298265749, + "loss": 0.119, + "num_input_tokens_seen": 13303200, + "step": 10255 + }, + { + "epoch": 0.5013070139007647, + "grad_norm": 0.003224806161597371, + "learning_rate": 0.2538866472084197, + "loss": 0.1338, + "num_input_tokens_seen": 13309504, + "step": 10260 + }, + { + "epoch": 0.501551315564459, + "grad_norm": 0.0019217501394450665, + "learning_rate": 0.25384414856963794, + "loss": 0.1514, + "num_input_tokens_seen": 13315712, + "step": 10265 + }, + { + "epoch": 0.5017956172281534, + "grad_norm": 0.0024106239434331656, + "learning_rate": 0.25380163391678356, + "loss": 0.1181, + "num_input_tokens_seen": 13322272, + "step": 10270 + }, + { + "epoch": 0.5020399188918476, + "grad_norm": 0.0010539956856518984, + "learning_rate": 0.2537591032564127, + "loss": 0.0945, + "num_input_tokens_seen": 13328416, + "step": 10275 + }, + { + "epoch": 0.502284220555542, + "grad_norm": 0.0014405617257580161, + "learning_rate": 0.25371655659508424, + "loss": 0.1104, + "num_input_tokens_seen": 13335296, + "step": 10280 + }, + { + "epoch": 0.5025285222192363, + "grad_norm": 0.002158868359401822, + "learning_rate": 0.25367399393935935, + "loss": 0.1092, + "num_input_tokens_seen": 13342496, + "step": 10285 + }, + { + "epoch": 0.5027728238829307, + "grad_norm": 0.0011546713067218661, + "learning_rate": 0.25363141529580174, + "loss": 0.1179, + "num_input_tokens_seen": 13348576, + "step": 10290 + }, + { + "epoch": 0.5030171255466249, + "grad_norm": 0.0014888821169734001, + "learning_rate": 0.2535888206709776, + "loss": 0.1196, + "num_input_tokens_seen": 13355008, + "step": 10295 + }, + { + "epoch": 0.5032614272103193, + "grad_norm": 0.0012215025490149856, + "learning_rate": 0.2535462100714555, + "loss": 0.1066, + "num_input_tokens_seen": 13361728, + "step": 10300 + }, + { + "epoch": 0.5035057288740137, + "grad_norm": 0.0014805456157773733, + "learning_rate": 0.2535035835038066, + "loss": 0.1004, + "num_input_tokens_seen": 13368032, + "step": 10305 + }, + { + "epoch": 0.503750030537708, + "grad_norm": 0.0020920278038829565, + "learning_rate": 0.2534609409746044, + "loss": 0.1404, + "num_input_tokens_seen": 13374752, + "step": 10310 + }, + { + "epoch": 0.5039943322014023, + "grad_norm": 0.001314655877649784, + "learning_rate": 0.253418282490425, + "loss": 0.1121, + "num_input_tokens_seen": 13380800, + "step": 10315 + }, + { + "epoch": 0.5042386338650966, + "grad_norm": 0.0011245508212596178, + "learning_rate": 0.2533756080578467, + "loss": 0.0992, + "num_input_tokens_seen": 13387200, + "step": 10320 + }, + { + "epoch": 0.504482935528791, + "grad_norm": 0.00118116638623178, + "learning_rate": 0.25333291768345056, + "loss": 0.1203, + "num_input_tokens_seen": 13393920, + "step": 10325 + }, + { + "epoch": 0.5047272371924852, + "grad_norm": 0.0016984047833830118, + "learning_rate": 0.25329021137381996, + "loss": 0.1008, + "num_input_tokens_seen": 13400736, + "step": 10330 + }, + { + "epoch": 0.5049715388561796, + "grad_norm": 0.0014488531742244959, + "learning_rate": 0.25324748913554074, + "loss": 0.1242, + "num_input_tokens_seen": 13407424, + "step": 10335 + }, + { + "epoch": 0.5052158405198739, + "grad_norm": 0.001806766726076603, + "learning_rate": 0.2532047509752013, + "loss": 0.1205, + "num_input_tokens_seen": 13414432, + "step": 10340 + }, + { + "epoch": 0.5054601421835683, + "grad_norm": 0.0016406466020271182, + "learning_rate": 0.25316199689939217, + "loss": 0.1217, + "num_input_tokens_seen": 13420960, + "step": 10345 + }, + { + "epoch": 0.5057044438472627, + "grad_norm": 0.0015145428478717804, + "learning_rate": 0.2531192269147068, + "loss": 0.1493, + "num_input_tokens_seen": 13427168, + "step": 10350 + }, + { + "epoch": 0.5059487455109569, + "grad_norm": 0.002183835953474045, + "learning_rate": 0.2530764410277407, + "loss": 0.0907, + "num_input_tokens_seen": 13433568, + "step": 10355 + }, + { + "epoch": 0.5061930471746513, + "grad_norm": 0.0009240178624168038, + "learning_rate": 0.25303363924509203, + "loss": 0.1119, + "num_input_tokens_seen": 13439648, + "step": 10360 + }, + { + "epoch": 0.5064373488383456, + "grad_norm": 0.001213107374496758, + "learning_rate": 0.25299082157336145, + "loss": 0.1051, + "num_input_tokens_seen": 13446272, + "step": 10365 + }, + { + "epoch": 0.50668165050204, + "grad_norm": 0.0012211932335048914, + "learning_rate": 0.2529479880191519, + "loss": 0.0955, + "num_input_tokens_seen": 13452672, + "step": 10370 + }, + { + "epoch": 0.5069259521657342, + "grad_norm": 0.0011349242413416505, + "learning_rate": 0.2529051385890689, + "loss": 0.1124, + "num_input_tokens_seen": 13459040, + "step": 10375 + }, + { + "epoch": 0.5071702538294286, + "grad_norm": 0.0013254669029265642, + "learning_rate": 0.2528622732897203, + "loss": 0.1109, + "num_input_tokens_seen": 13465824, + "step": 10380 + }, + { + "epoch": 0.5074145554931229, + "grad_norm": 0.0012858441332355142, + "learning_rate": 0.25281939212771654, + "loss": 0.1272, + "num_input_tokens_seen": 13471904, + "step": 10385 + }, + { + "epoch": 0.5076588571568172, + "grad_norm": 0.0025245589204132557, + "learning_rate": 0.2527764951096704, + "loss": 0.1172, + "num_input_tokens_seen": 13477856, + "step": 10390 + }, + { + "epoch": 0.5079031588205115, + "grad_norm": 0.002354786265641451, + "learning_rate": 0.2527335822421971, + "loss": 0.1291, + "num_input_tokens_seen": 13483552, + "step": 10395 + }, + { + "epoch": 0.5081474604842059, + "grad_norm": 0.001969563076272607, + "learning_rate": 0.25269065353191444, + "loss": 0.0973, + "num_input_tokens_seen": 13489760, + "step": 10400 + }, + { + "epoch": 0.5081474604842059, + "eval_loss": 0.11623179167509079, + "eval_runtime": 402.3775, + "eval_samples_per_second": 90.425, + "eval_steps_per_second": 22.608, + "num_input_tokens_seen": 13489760, + "step": 10400 + }, + { + "epoch": 0.5083917621479003, + "grad_norm": 0.001495670061558485, + "learning_rate": 0.2526477089854425, + "loss": 0.1111, + "num_input_tokens_seen": 13496160, + "step": 10405 + }, + { + "epoch": 0.5086360638115945, + "grad_norm": 0.002736422698944807, + "learning_rate": 0.25260474860940385, + "loss": 0.0747, + "num_input_tokens_seen": 13503072, + "step": 10410 + }, + { + "epoch": 0.5088803654752889, + "grad_norm": 0.0023960794787853956, + "learning_rate": 0.2525617724104236, + "loss": 0.1207, + "num_input_tokens_seen": 13509120, + "step": 10415 + }, + { + "epoch": 0.5091246671389832, + "grad_norm": 0.001421345747075975, + "learning_rate": 0.25251878039512915, + "loss": 0.1026, + "num_input_tokens_seen": 13515616, + "step": 10420 + }, + { + "epoch": 0.5093689688026776, + "grad_norm": 0.0021058539859950542, + "learning_rate": 0.25247577257015047, + "loss": 0.1254, + "num_input_tokens_seen": 13522240, + "step": 10425 + }, + { + "epoch": 0.5096132704663718, + "grad_norm": 0.0012945587513968349, + "learning_rate": 0.2524327489421198, + "loss": 0.098, + "num_input_tokens_seen": 13528512, + "step": 10430 + }, + { + "epoch": 0.5098575721300662, + "grad_norm": 0.0019859240856021643, + "learning_rate": 0.25238970951767203, + "loss": 0.1364, + "num_input_tokens_seen": 13534528, + "step": 10435 + }, + { + "epoch": 0.5101018737937605, + "grad_norm": 0.001702063367702067, + "learning_rate": 0.25234665430344433, + "loss": 0.0931, + "num_input_tokens_seen": 13541120, + "step": 10440 + }, + { + "epoch": 0.5103461754574549, + "grad_norm": 0.001672688638791442, + "learning_rate": 0.2523035833060764, + "loss": 0.1293, + "num_input_tokens_seen": 13547904, + "step": 10445 + }, + { + "epoch": 0.5105904771211492, + "grad_norm": 0.001304520876146853, + "learning_rate": 0.2522604965322103, + "loss": 0.1048, + "num_input_tokens_seen": 13554400, + "step": 10450 + }, + { + "epoch": 0.5108347787848435, + "grad_norm": 0.0019980366341769695, + "learning_rate": 0.25221739398849047, + "loss": 0.1181, + "num_input_tokens_seen": 13560864, + "step": 10455 + }, + { + "epoch": 0.5110790804485379, + "grad_norm": 0.001317598856985569, + "learning_rate": 0.252174275681564, + "loss": 0.1105, + "num_input_tokens_seen": 13567264, + "step": 10460 + }, + { + "epoch": 0.5113233821122322, + "grad_norm": 0.002019987441599369, + "learning_rate": 0.2521311416180802, + "loss": 0.1086, + "num_input_tokens_seen": 13573568, + "step": 10465 + }, + { + "epoch": 0.5115676837759265, + "grad_norm": 0.001819878350943327, + "learning_rate": 0.25208799180469094, + "loss": 0.1044, + "num_input_tokens_seen": 13579552, + "step": 10470 + }, + { + "epoch": 0.5118119854396208, + "grad_norm": 0.0014593073865398765, + "learning_rate": 0.2520448262480504, + "loss": 0.0972, + "num_input_tokens_seen": 13586080, + "step": 10475 + }, + { + "epoch": 0.5120562871033152, + "grad_norm": 0.0013555069454014301, + "learning_rate": 0.25200164495481525, + "loss": 0.0876, + "num_input_tokens_seen": 13592928, + "step": 10480 + }, + { + "epoch": 0.5123005887670095, + "grad_norm": 0.003069641999900341, + "learning_rate": 0.25195844793164474, + "loss": 0.1102, + "num_input_tokens_seen": 13600064, + "step": 10485 + }, + { + "epoch": 0.5125448904307038, + "grad_norm": 0.00222374708391726, + "learning_rate": 0.2519152351852001, + "loss": 0.1215, + "num_input_tokens_seen": 13606880, + "step": 10490 + }, + { + "epoch": 0.5127891920943982, + "grad_norm": 0.001463662483729422, + "learning_rate": 0.25187200672214555, + "loss": 0.1189, + "num_input_tokens_seen": 13613216, + "step": 10495 + }, + { + "epoch": 0.5130334937580925, + "grad_norm": 0.0009306384017691016, + "learning_rate": 0.2518287625491473, + "loss": 0.1015, + "num_input_tokens_seen": 13619776, + "step": 10500 + }, + { + "epoch": 0.5132777954217869, + "grad_norm": 0.0016621758695691824, + "learning_rate": 0.25178550267287425, + "loss": 0.0911, + "num_input_tokens_seen": 13626400, + "step": 10505 + }, + { + "epoch": 0.5135220970854811, + "grad_norm": 0.0022571177687495947, + "learning_rate": 0.2517422270999976, + "loss": 0.1117, + "num_input_tokens_seen": 13632480, + "step": 10510 + }, + { + "epoch": 0.5137663987491755, + "grad_norm": 0.0017553361831232905, + "learning_rate": 0.2516989358371909, + "loss": 0.1222, + "num_input_tokens_seen": 13639456, + "step": 10515 + }, + { + "epoch": 0.5140107004128698, + "grad_norm": 0.001829514279961586, + "learning_rate": 0.25165562889113025, + "loss": 0.1166, + "num_input_tokens_seen": 13645600, + "step": 10520 + }, + { + "epoch": 0.5142550020765642, + "grad_norm": 0.0012940785381942987, + "learning_rate": 0.2516123062684942, + "loss": 0.0922, + "num_input_tokens_seen": 13651904, + "step": 10525 + }, + { + "epoch": 0.5144993037402584, + "grad_norm": 0.001664554001763463, + "learning_rate": 0.25156896797596356, + "loss": 0.1141, + "num_input_tokens_seen": 13658336, + "step": 10530 + }, + { + "epoch": 0.5147436054039528, + "grad_norm": 0.0018766829743981361, + "learning_rate": 0.2515256140202216, + "loss": 0.1066, + "num_input_tokens_seen": 13664480, + "step": 10535 + }, + { + "epoch": 0.5149879070676471, + "grad_norm": 0.0016662696143612266, + "learning_rate": 0.25148224440795425, + "loss": 0.1188, + "num_input_tokens_seen": 13670720, + "step": 10540 + }, + { + "epoch": 0.5152322087313415, + "grad_norm": 0.001529004774056375, + "learning_rate": 0.2514388591458494, + "loss": 0.1067, + "num_input_tokens_seen": 13677152, + "step": 10545 + }, + { + "epoch": 0.5154765103950358, + "grad_norm": 0.001983480527997017, + "learning_rate": 0.2513954582405977, + "loss": 0.1396, + "num_input_tokens_seen": 13683616, + "step": 10550 + }, + { + "epoch": 0.5157208120587301, + "grad_norm": 0.0013539961073547602, + "learning_rate": 0.2513520416988922, + "loss": 0.1128, + "num_input_tokens_seen": 13689696, + "step": 10555 + }, + { + "epoch": 0.5159651137224245, + "grad_norm": 0.0009956650901585817, + "learning_rate": 0.2513086095274281, + "loss": 0.1248, + "num_input_tokens_seen": 13696416, + "step": 10560 + }, + { + "epoch": 0.5162094153861188, + "grad_norm": 0.0012999052414670587, + "learning_rate": 0.25126516173290336, + "loss": 0.1023, + "num_input_tokens_seen": 13703136, + "step": 10565 + }, + { + "epoch": 0.5164537170498131, + "grad_norm": 0.001188030350022018, + "learning_rate": 0.2512216983220181, + "loss": 0.0934, + "num_input_tokens_seen": 13709216, + "step": 10570 + }, + { + "epoch": 0.5166980187135074, + "grad_norm": 0.0012600376503542066, + "learning_rate": 0.25117821930147494, + "loss": 0.0994, + "num_input_tokens_seen": 13716480, + "step": 10575 + }, + { + "epoch": 0.5169423203772018, + "grad_norm": 0.0017034104093909264, + "learning_rate": 0.2511347246779788, + "loss": 0.0962, + "num_input_tokens_seen": 13723936, + "step": 10580 + }, + { + "epoch": 0.517186622040896, + "grad_norm": 0.001125931041315198, + "learning_rate": 0.25109121445823723, + "loss": 0.1147, + "num_input_tokens_seen": 13730592, + "step": 10585 + }, + { + "epoch": 0.5174309237045904, + "grad_norm": 0.0026348368264734745, + "learning_rate": 0.25104768864896004, + "loss": 0.1201, + "num_input_tokens_seen": 13737152, + "step": 10590 + }, + { + "epoch": 0.5176752253682848, + "grad_norm": 0.0017300941981375217, + "learning_rate": 0.2510041472568594, + "loss": 0.1035, + "num_input_tokens_seen": 13743808, + "step": 10595 + }, + { + "epoch": 0.5179195270319791, + "grad_norm": 0.0014441937673836946, + "learning_rate": 0.25096059028864987, + "loss": 0.1063, + "num_input_tokens_seen": 13750592, + "step": 10600 + }, + { + "epoch": 0.5179195270319791, + "eval_loss": 0.11015040427446365, + "eval_runtime": 402.01, + "eval_samples_per_second": 90.508, + "eval_steps_per_second": 22.629, + "num_input_tokens_seen": 13750592, + "step": 10600 + }, + { + "epoch": 0.5181638286956735, + "grad_norm": 0.0019387531792744994, + "learning_rate": 0.25091701775104863, + "loss": 0.111, + "num_input_tokens_seen": 13756896, + "step": 10605 + }, + { + "epoch": 0.5184081303593677, + "grad_norm": 0.0020773683208972216, + "learning_rate": 0.250873429650775, + "loss": 0.1225, + "num_input_tokens_seen": 13763328, + "step": 10610 + }, + { + "epoch": 0.5186524320230621, + "grad_norm": 0.0013712659711018205, + "learning_rate": 0.25082982599455095, + "loss": 0.1017, + "num_input_tokens_seen": 13769696, + "step": 10615 + }, + { + "epoch": 0.5188967336867564, + "grad_norm": 0.0012630640994757414, + "learning_rate": 0.2507862067891006, + "loss": 0.1074, + "num_input_tokens_seen": 13775808, + "step": 10620 + }, + { + "epoch": 0.5191410353504508, + "grad_norm": 0.0019027084344998002, + "learning_rate": 0.25074257204115064, + "loss": 0.1164, + "num_input_tokens_seen": 13781952, + "step": 10625 + }, + { + "epoch": 0.519385337014145, + "grad_norm": 0.0013097012415528297, + "learning_rate": 0.25069892175742997, + "loss": 0.1093, + "num_input_tokens_seen": 13788064, + "step": 10630 + }, + { + "epoch": 0.5196296386778394, + "grad_norm": 0.0021589817479252815, + "learning_rate": 0.25065525594467014, + "loss": 0.1172, + "num_input_tokens_seen": 13794112, + "step": 10635 + }, + { + "epoch": 0.5198739403415338, + "grad_norm": 0.0017306477529928088, + "learning_rate": 0.2506115746096049, + "loss": 0.1229, + "num_input_tokens_seen": 13800736, + "step": 10640 + }, + { + "epoch": 0.520118242005228, + "grad_norm": 0.0012866974575445056, + "learning_rate": 0.25056787775897055, + "loss": 0.1022, + "num_input_tokens_seen": 13807008, + "step": 10645 + }, + { + "epoch": 0.5203625436689224, + "grad_norm": 0.0009840653510764241, + "learning_rate": 0.2505241653995056, + "loss": 0.1145, + "num_input_tokens_seen": 13813760, + "step": 10650 + }, + { + "epoch": 0.5206068453326167, + "grad_norm": 0.001333807478658855, + "learning_rate": 0.25048043753795113, + "loss": 0.093, + "num_input_tokens_seen": 13820224, + "step": 10655 + }, + { + "epoch": 0.5208511469963111, + "grad_norm": 0.0014747707173228264, + "learning_rate": 0.2504366941810504, + "loss": 0.135, + "num_input_tokens_seen": 13826976, + "step": 10660 + }, + { + "epoch": 0.5210954486600053, + "grad_norm": 0.0013075354509055614, + "learning_rate": 0.2503929353355493, + "loss": 0.113, + "num_input_tokens_seen": 13833184, + "step": 10665 + }, + { + "epoch": 0.5213397503236997, + "grad_norm": 0.0019209507154300809, + "learning_rate": 0.250349161008196, + "loss": 0.1037, + "num_input_tokens_seen": 13839936, + "step": 10670 + }, + { + "epoch": 0.521584051987394, + "grad_norm": 0.0012993625132367015, + "learning_rate": 0.2503053712057409, + "loss": 0.0994, + "num_input_tokens_seen": 13846144, + "step": 10675 + }, + { + "epoch": 0.5218283536510884, + "grad_norm": 0.002098111668601632, + "learning_rate": 0.25026156593493715, + "loss": 0.099, + "num_input_tokens_seen": 13852736, + "step": 10680 + }, + { + "epoch": 0.5220726553147826, + "grad_norm": 0.0019454223802313209, + "learning_rate": 0.2502177452025399, + "loss": 0.1076, + "num_input_tokens_seen": 13859552, + "step": 10685 + }, + { + "epoch": 0.522316956978477, + "grad_norm": 0.0017071864567697048, + "learning_rate": 0.25017390901530695, + "loss": 0.0904, + "num_input_tokens_seen": 13866016, + "step": 10690 + }, + { + "epoch": 0.5225612586421714, + "grad_norm": 0.0030082897283136845, + "learning_rate": 0.2501300573799984, + "loss": 0.1092, + "num_input_tokens_seen": 13872160, + "step": 10695 + }, + { + "epoch": 0.5228055603058657, + "grad_norm": 0.002473636996001005, + "learning_rate": 0.2500861903033766, + "loss": 0.1271, + "num_input_tokens_seen": 13878368, + "step": 10700 + }, + { + "epoch": 0.52304986196956, + "grad_norm": 0.0015367644373327494, + "learning_rate": 0.25004230779220654, + "loss": 0.1107, + "num_input_tokens_seen": 13884928, + "step": 10705 + }, + { + "epoch": 0.5232941636332543, + "grad_norm": 0.001264012767933309, + "learning_rate": 0.24999840985325542, + "loss": 0.1134, + "num_input_tokens_seen": 13891360, + "step": 10710 + }, + { + "epoch": 0.5235384652969487, + "grad_norm": 0.0014951216289773583, + "learning_rate": 0.24995449649329285, + "loss": 0.1135, + "num_input_tokens_seen": 13898272, + "step": 10715 + }, + { + "epoch": 0.523782766960643, + "grad_norm": 0.001311751315370202, + "learning_rate": 0.2499105677190908, + "loss": 0.1165, + "num_input_tokens_seen": 13904992, + "step": 10720 + }, + { + "epoch": 0.5240270686243373, + "grad_norm": 0.001982693560421467, + "learning_rate": 0.24986662353742364, + "loss": 0.0999, + "num_input_tokens_seen": 13911744, + "step": 10725 + }, + { + "epoch": 0.5242713702880316, + "grad_norm": 0.0020960066467523575, + "learning_rate": 0.24982266395506814, + "loss": 0.1084, + "num_input_tokens_seen": 13918112, + "step": 10730 + }, + { + "epoch": 0.524515671951726, + "grad_norm": 0.001956281950697303, + "learning_rate": 0.2497786889788034, + "loss": 0.1236, + "num_input_tokens_seen": 13924224, + "step": 10735 + }, + { + "epoch": 0.5247599736154204, + "grad_norm": 0.001759157283231616, + "learning_rate": 0.24973469861541095, + "loss": 0.1274, + "num_input_tokens_seen": 13930624, + "step": 10740 + }, + { + "epoch": 0.5250042752791146, + "grad_norm": 0.0009564183419570327, + "learning_rate": 0.24969069287167456, + "loss": 0.104, + "num_input_tokens_seen": 13936992, + "step": 10745 + }, + { + "epoch": 0.525248576942809, + "grad_norm": 0.001706930692307651, + "learning_rate": 0.2496466717543806, + "loss": 0.1418, + "num_input_tokens_seen": 13943360, + "step": 10750 + }, + { + "epoch": 0.5254928786065033, + "grad_norm": 0.001745807589031756, + "learning_rate": 0.24960263527031762, + "loss": 0.1238, + "num_input_tokens_seen": 13949792, + "step": 10755 + }, + { + "epoch": 0.5257371802701977, + "grad_norm": 0.0031445471104234457, + "learning_rate": 0.24955858342627657, + "loss": 0.1247, + "num_input_tokens_seen": 13956480, + "step": 10760 + }, + { + "epoch": 0.5259814819338919, + "grad_norm": 0.0013161319075152278, + "learning_rate": 0.24951451622905083, + "loss": 0.1017, + "num_input_tokens_seen": 13962688, + "step": 10765 + }, + { + "epoch": 0.5262257835975863, + "grad_norm": 0.0013531798031181097, + "learning_rate": 0.24947043368543612, + "loss": 0.0987, + "num_input_tokens_seen": 13969888, + "step": 10770 + }, + { + "epoch": 0.5264700852612806, + "grad_norm": 0.0017245005583390594, + "learning_rate": 0.2494263358022305, + "loss": 0.0956, + "num_input_tokens_seen": 13976768, + "step": 10775 + }, + { + "epoch": 0.526714386924975, + "grad_norm": 0.001760319690220058, + "learning_rate": 0.24938222258623444, + "loss": 0.1448, + "num_input_tokens_seen": 13983360, + "step": 10780 + }, + { + "epoch": 0.5269586885886693, + "grad_norm": 0.0015171509003266692, + "learning_rate": 0.24933809404425075, + "loss": 0.1086, + "num_input_tokens_seen": 13989504, + "step": 10785 + }, + { + "epoch": 0.5272029902523636, + "grad_norm": 0.00176490587182343, + "learning_rate": 0.24929395018308453, + "loss": 0.1128, + "num_input_tokens_seen": 13996320, + "step": 10790 + }, + { + "epoch": 0.527447291916058, + "grad_norm": 0.0024009624030441046, + "learning_rate": 0.24924979100954348, + "loss": 0.1024, + "num_input_tokens_seen": 14003040, + "step": 10795 + }, + { + "epoch": 0.5276915935797523, + "grad_norm": 0.001639460795558989, + "learning_rate": 0.24920561653043735, + "loss": 0.1189, + "num_input_tokens_seen": 14009088, + "step": 10800 + }, + { + "epoch": 0.5276915935797523, + "eval_loss": 0.11233407258987427, + "eval_runtime": 402.3578, + "eval_samples_per_second": 90.429, + "eval_steps_per_second": 22.609, + "num_input_tokens_seen": 14009088, + "step": 10800 + }, + { + "epoch": 0.5279358952434466, + "grad_norm": 0.0010506178950890899, + "learning_rate": 0.24916142675257846, + "loss": 0.1081, + "num_input_tokens_seen": 14015776, + "step": 10805 + }, + { + "epoch": 0.5281801969071409, + "grad_norm": 0.0010298562701791525, + "learning_rate": 0.24911722168278144, + "loss": 0.0939, + "num_input_tokens_seen": 14022144, + "step": 10810 + }, + { + "epoch": 0.5284244985708353, + "grad_norm": 0.0012871791841462255, + "learning_rate": 0.24907300132786328, + "loss": 0.1043, + "num_input_tokens_seen": 14028864, + "step": 10815 + }, + { + "epoch": 0.5286688002345296, + "grad_norm": 0.00202836561948061, + "learning_rate": 0.24902876569464322, + "loss": 0.1087, + "num_input_tokens_seen": 14035392, + "step": 10820 + }, + { + "epoch": 0.5289131018982239, + "grad_norm": 0.001338908914476633, + "learning_rate": 0.24898451478994305, + "loss": 0.0947, + "num_input_tokens_seen": 14041888, + "step": 10825 + }, + { + "epoch": 0.5291574035619182, + "grad_norm": 0.0018045541364699602, + "learning_rate": 0.2489402486205868, + "loss": 0.124, + "num_input_tokens_seen": 14048768, + "step": 10830 + }, + { + "epoch": 0.5294017052256126, + "grad_norm": 0.0019813550170511007, + "learning_rate": 0.24889596719340085, + "loss": 0.127, + "num_input_tokens_seen": 14054944, + "step": 10835 + }, + { + "epoch": 0.529646006889307, + "grad_norm": 0.0022456333972513676, + "learning_rate": 0.24885167051521392, + "loss": 0.108, + "num_input_tokens_seen": 14061568, + "step": 10840 + }, + { + "epoch": 0.5298903085530012, + "grad_norm": 0.0010676010278984904, + "learning_rate": 0.24880735859285716, + "loss": 0.0907, + "num_input_tokens_seen": 14067872, + "step": 10845 + }, + { + "epoch": 0.5301346102166956, + "grad_norm": 0.0007500868523493409, + "learning_rate": 0.24876303143316406, + "loss": 0.0966, + "num_input_tokens_seen": 14073824, + "step": 10850 + }, + { + "epoch": 0.5303789118803899, + "grad_norm": 0.0015099674928933382, + "learning_rate": 0.24871868904297031, + "loss": 0.111, + "num_input_tokens_seen": 14079968, + "step": 10855 + }, + { + "epoch": 0.5306232135440843, + "grad_norm": 0.002198414411395788, + "learning_rate": 0.24867433142911416, + "loss": 0.0933, + "num_input_tokens_seen": 14087040, + "step": 10860 + }, + { + "epoch": 0.5308675152077785, + "grad_norm": 0.0014415087644010782, + "learning_rate": 0.24862995859843612, + "loss": 0.0999, + "num_input_tokens_seen": 14094432, + "step": 10865 + }, + { + "epoch": 0.5311118168714729, + "grad_norm": 0.0021472019143402576, + "learning_rate": 0.24858557055777897, + "loss": 0.1329, + "num_input_tokens_seen": 14100480, + "step": 10870 + }, + { + "epoch": 0.5313561185351672, + "grad_norm": 0.0032043657265603542, + "learning_rate": 0.24854116731398793, + "loss": 0.0929, + "num_input_tokens_seen": 14106848, + "step": 10875 + }, + { + "epoch": 0.5316004201988616, + "grad_norm": 0.002146227518096566, + "learning_rate": 0.24849674887391052, + "loss": 0.0993, + "num_input_tokens_seen": 14112800, + "step": 10880 + }, + { + "epoch": 0.5318447218625559, + "grad_norm": 0.0020355740562081337, + "learning_rate": 0.2484523152443967, + "loss": 0.1238, + "num_input_tokens_seen": 14119328, + "step": 10885 + }, + { + "epoch": 0.5320890235262502, + "grad_norm": 0.002448880113661289, + "learning_rate": 0.24840786643229862, + "loss": 0.1341, + "num_input_tokens_seen": 14126080, + "step": 10890 + }, + { + "epoch": 0.5323333251899446, + "grad_norm": 0.001217106357216835, + "learning_rate": 0.2483634024444709, + "loss": 0.1217, + "num_input_tokens_seen": 14132320, + "step": 10895 + }, + { + "epoch": 0.5325776268536389, + "grad_norm": 0.0008433673647232354, + "learning_rate": 0.24831892328777033, + "loss": 0.107, + "num_input_tokens_seen": 14138464, + "step": 10900 + }, + { + "epoch": 0.5328219285173332, + "grad_norm": 0.001392542035318911, + "learning_rate": 0.2482744289690563, + "loss": 0.0955, + "num_input_tokens_seen": 14144768, + "step": 10905 + }, + { + "epoch": 0.5330662301810275, + "grad_norm": 0.0016061124624684453, + "learning_rate": 0.2482299194951903, + "loss": 0.1114, + "num_input_tokens_seen": 14151584, + "step": 10910 + }, + { + "epoch": 0.5333105318447219, + "grad_norm": 0.0012232520384714007, + "learning_rate": 0.2481853948730363, + "loss": 0.1185, + "num_input_tokens_seen": 14158400, + "step": 10915 + }, + { + "epoch": 0.5335548335084161, + "grad_norm": 0.0012601087801158428, + "learning_rate": 0.24814085510946052, + "loss": 0.1115, + "num_input_tokens_seen": 14164608, + "step": 10920 + }, + { + "epoch": 0.5337991351721105, + "grad_norm": 0.0006942101754248142, + "learning_rate": 0.24809630021133158, + "loss": 0.0972, + "num_input_tokens_seen": 14171232, + "step": 10925 + }, + { + "epoch": 0.5340434368358048, + "grad_norm": 0.0011418143985792994, + "learning_rate": 0.24805173018552037, + "loss": 0.105, + "num_input_tokens_seen": 14177696, + "step": 10930 + }, + { + "epoch": 0.5342877384994992, + "grad_norm": 0.0013452030252665281, + "learning_rate": 0.2480071450389002, + "loss": 0.1159, + "num_input_tokens_seen": 14184544, + "step": 10935 + }, + { + "epoch": 0.5345320401631936, + "grad_norm": 0.0008317091269418597, + "learning_rate": 0.24796254477834662, + "loss": 0.0802, + "num_input_tokens_seen": 14190784, + "step": 10940 + }, + { + "epoch": 0.5347763418268878, + "grad_norm": 0.0008900338434614241, + "learning_rate": 0.24791792941073754, + "loss": 0.1174, + "num_input_tokens_seen": 14196896, + "step": 10945 + }, + { + "epoch": 0.5350206434905822, + "grad_norm": 0.0018141018226742744, + "learning_rate": 0.2478732989429533, + "loss": 0.1042, + "num_input_tokens_seen": 14203424, + "step": 10950 + }, + { + "epoch": 0.5352649451542765, + "grad_norm": 0.001401800662279129, + "learning_rate": 0.24782865338187632, + "loss": 0.0926, + "num_input_tokens_seen": 14210080, + "step": 10955 + }, + { + "epoch": 0.5355092468179709, + "grad_norm": 0.0021286909468472004, + "learning_rate": 0.2477839927343916, + "loss": 0.1395, + "num_input_tokens_seen": 14216160, + "step": 10960 + }, + { + "epoch": 0.5357535484816651, + "grad_norm": 0.0017903412226587534, + "learning_rate": 0.2477393170073864, + "loss": 0.1217, + "num_input_tokens_seen": 14222592, + "step": 10965 + }, + { + "epoch": 0.5359978501453595, + "grad_norm": 0.0031504356302320957, + "learning_rate": 0.2476946262077503, + "loss": 0.0814, + "num_input_tokens_seen": 14229664, + "step": 10970 + }, + { + "epoch": 0.5362421518090538, + "grad_norm": 0.000762486073654145, + "learning_rate": 0.24764992034237507, + "loss": 0.1172, + "num_input_tokens_seen": 14235968, + "step": 10975 + }, + { + "epoch": 0.5364864534727481, + "grad_norm": 0.0021405688021332026, + "learning_rate": 0.24760519941815498, + "loss": 0.1046, + "num_input_tokens_seen": 14242592, + "step": 10980 + }, + { + "epoch": 0.5367307551364425, + "grad_norm": 0.001193955889903009, + "learning_rate": 0.2475604634419866, + "loss": 0.1158, + "num_input_tokens_seen": 14249056, + "step": 10985 + }, + { + "epoch": 0.5369750568001368, + "grad_norm": 0.0013813746627420187, + "learning_rate": 0.24751571242076872, + "loss": 0.1162, + "num_input_tokens_seen": 14255488, + "step": 10990 + }, + { + "epoch": 0.5372193584638312, + "grad_norm": 0.001454840530641377, + "learning_rate": 0.2474709463614025, + "loss": 0.1134, + "num_input_tokens_seen": 14262080, + "step": 10995 + }, + { + "epoch": 0.5374636601275254, + "grad_norm": 0.0015893186209723353, + "learning_rate": 0.24742616527079145, + "loss": 0.1164, + "num_input_tokens_seen": 14268352, + "step": 11000 + }, + { + "epoch": 0.5374636601275254, + "eval_loss": 0.11256943643093109, + "eval_runtime": 402.7188, + "eval_samples_per_second": 90.348, + "eval_steps_per_second": 22.589, + "num_input_tokens_seen": 14268352, + "step": 11000 + }, + { + "epoch": 0.5377079617912198, + "grad_norm": 0.0013034285511821508, + "learning_rate": 0.24738136915584139, + "loss": 0.083, + "num_input_tokens_seen": 14274656, + "step": 11005 + }, + { + "epoch": 0.5379522634549141, + "grad_norm": 0.0026484366971999407, + "learning_rate": 0.24733655802346047, + "loss": 0.1336, + "num_input_tokens_seen": 14280672, + "step": 11010 + }, + { + "epoch": 0.5381965651186085, + "grad_norm": 0.002655264688655734, + "learning_rate": 0.24729173188055906, + "loss": 0.1374, + "num_input_tokens_seen": 14287104, + "step": 11015 + }, + { + "epoch": 0.5384408667823027, + "grad_norm": 0.002454804489389062, + "learning_rate": 0.24724689073404996, + "loss": 0.1113, + "num_input_tokens_seen": 14293504, + "step": 11020 + }, + { + "epoch": 0.5386851684459971, + "grad_norm": 0.0016575338086113334, + "learning_rate": 0.24720203459084822, + "loss": 0.1205, + "num_input_tokens_seen": 14299328, + "step": 11025 + }, + { + "epoch": 0.5389294701096915, + "grad_norm": 0.0013502378715202212, + "learning_rate": 0.24715716345787123, + "loss": 0.0914, + "num_input_tokens_seen": 14305664, + "step": 11030 + }, + { + "epoch": 0.5391737717733858, + "grad_norm": 0.0017488976009190083, + "learning_rate": 0.2471122773420387, + "loss": 0.0949, + "num_input_tokens_seen": 14312096, + "step": 11035 + }, + { + "epoch": 0.5394180734370801, + "grad_norm": 0.001063516829162836, + "learning_rate": 0.24706737625027259, + "loss": 0.0861, + "num_input_tokens_seen": 14318368, + "step": 11040 + }, + { + "epoch": 0.5396623751007744, + "grad_norm": 0.0013701067073270679, + "learning_rate": 0.24702246018949725, + "loss": 0.1113, + "num_input_tokens_seen": 14324448, + "step": 11045 + }, + { + "epoch": 0.5399066767644688, + "grad_norm": 0.0030957504641264677, + "learning_rate": 0.2469775291666393, + "loss": 0.11, + "num_input_tokens_seen": 14331200, + "step": 11050 + }, + { + "epoch": 0.5401509784281631, + "grad_norm": 0.0013287820620462298, + "learning_rate": 0.24693258318862765, + "loss": 0.1144, + "num_input_tokens_seen": 14337472, + "step": 11055 + }, + { + "epoch": 0.5403952800918574, + "grad_norm": 0.001855275477282703, + "learning_rate": 0.2468876222623935, + "loss": 0.1032, + "num_input_tokens_seen": 14343872, + "step": 11060 + }, + { + "epoch": 0.5406395817555517, + "grad_norm": 0.001819952973164618, + "learning_rate": 0.2468426463948705, + "loss": 0.1029, + "num_input_tokens_seen": 14350720, + "step": 11065 + }, + { + "epoch": 0.5408838834192461, + "grad_norm": 0.0017901096725836396, + "learning_rate": 0.24679765559299438, + "loss": 0.112, + "num_input_tokens_seen": 14357280, + "step": 11070 + }, + { + "epoch": 0.5411281850829404, + "grad_norm": 0.0012608241522684693, + "learning_rate": 0.24675264986370332, + "loss": 0.1029, + "num_input_tokens_seen": 14363872, + "step": 11075 + }, + { + "epoch": 0.5413724867466347, + "grad_norm": 0.0018643528455868363, + "learning_rate": 0.2467076292139378, + "loss": 0.1379, + "num_input_tokens_seen": 14370048, + "step": 11080 + }, + { + "epoch": 0.5416167884103291, + "grad_norm": 0.001080165384337306, + "learning_rate": 0.24666259365064055, + "loss": 0.1101, + "num_input_tokens_seen": 14376416, + "step": 11085 + }, + { + "epoch": 0.5418610900740234, + "grad_norm": 0.0017822706140577793, + "learning_rate": 0.24661754318075663, + "loss": 0.1176, + "num_input_tokens_seen": 14382816, + "step": 11090 + }, + { + "epoch": 0.5421053917377178, + "grad_norm": 0.001504267449490726, + "learning_rate": 0.2465724778112334, + "loss": 0.1052, + "num_input_tokens_seen": 14389888, + "step": 11095 + }, + { + "epoch": 0.542349693401412, + "grad_norm": 0.0011343329679220915, + "learning_rate": 0.24652739754902042, + "loss": 0.1002, + "num_input_tokens_seen": 14396448, + "step": 11100 + }, + { + "epoch": 0.5425939950651064, + "grad_norm": 0.0014250566018745303, + "learning_rate": 0.24648230240106975, + "loss": 0.1278, + "num_input_tokens_seen": 14402688, + "step": 11105 + }, + { + "epoch": 0.5428382967288007, + "grad_norm": 0.001798174693249166, + "learning_rate": 0.2464371923743356, + "loss": 0.1035, + "num_input_tokens_seen": 14409472, + "step": 11110 + }, + { + "epoch": 0.5430825983924951, + "grad_norm": 0.0007593840709887445, + "learning_rate": 0.24639206747577444, + "loss": 0.1099, + "num_input_tokens_seen": 14415968, + "step": 11115 + }, + { + "epoch": 0.5433269000561893, + "grad_norm": 0.0013187275035306811, + "learning_rate": 0.24634692771234515, + "loss": 0.0948, + "num_input_tokens_seen": 14422688, + "step": 11120 + }, + { + "epoch": 0.5435712017198837, + "grad_norm": 0.0011279114987701178, + "learning_rate": 0.2463017730910088, + "loss": 0.1152, + "num_input_tokens_seen": 14429312, + "step": 11125 + }, + { + "epoch": 0.5438155033835781, + "grad_norm": 0.001235799863934517, + "learning_rate": 0.2462566036187289, + "loss": 0.0853, + "num_input_tokens_seen": 14435968, + "step": 11130 + }, + { + "epoch": 0.5440598050472724, + "grad_norm": 0.001079428126104176, + "learning_rate": 0.24621141930247106, + "loss": 0.1147, + "num_input_tokens_seen": 14442592, + "step": 11135 + }, + { + "epoch": 0.5443041067109667, + "grad_norm": 0.001784582156687975, + "learning_rate": 0.2461662201492033, + "loss": 0.1179, + "num_input_tokens_seen": 14448832, + "step": 11140 + }, + { + "epoch": 0.544548408374661, + "grad_norm": 0.0017814160091802478, + "learning_rate": 0.24612100616589586, + "loss": 0.1184, + "num_input_tokens_seen": 14455328, + "step": 11145 + }, + { + "epoch": 0.5447927100383554, + "grad_norm": 0.0009529789676889777, + "learning_rate": 0.24607577735952135, + "loss": 0.0932, + "num_input_tokens_seen": 14461664, + "step": 11150 + }, + { + "epoch": 0.5450370117020497, + "grad_norm": 0.001968234544619918, + "learning_rate": 0.24603053373705464, + "loss": 0.1336, + "num_input_tokens_seen": 14467968, + "step": 11155 + }, + { + "epoch": 0.545281313365744, + "grad_norm": 0.0013042062055319548, + "learning_rate": 0.2459852753054728, + "loss": 0.1012, + "num_input_tokens_seen": 14474368, + "step": 11160 + }, + { + "epoch": 0.5455256150294383, + "grad_norm": 0.0013361782766878605, + "learning_rate": 0.24594000207175526, + "loss": 0.094, + "num_input_tokens_seen": 14481216, + "step": 11165 + }, + { + "epoch": 0.5457699166931327, + "grad_norm": 0.0014705279609188437, + "learning_rate": 0.2458947140428838, + "loss": 0.1054, + "num_input_tokens_seen": 14487776, + "step": 11170 + }, + { + "epoch": 0.5460142183568271, + "grad_norm": 0.0015681280056014657, + "learning_rate": 0.24584941122584233, + "loss": 0.1037, + "num_input_tokens_seen": 14494304, + "step": 11175 + }, + { + "epoch": 0.5462585200205213, + "grad_norm": 0.001394380466081202, + "learning_rate": 0.24580409362761713, + "loss": 0.0942, + "num_input_tokens_seen": 14501120, + "step": 11180 + }, + { + "epoch": 0.5465028216842157, + "grad_norm": 0.0019721011631190777, + "learning_rate": 0.2457587612551967, + "loss": 0.1327, + "num_input_tokens_seen": 14507424, + "step": 11185 + }, + { + "epoch": 0.54674712334791, + "grad_norm": 0.0013357134303078055, + "learning_rate": 0.24571341411557193, + "loss": 0.1078, + "num_input_tokens_seen": 14513728, + "step": 11190 + }, + { + "epoch": 0.5469914250116044, + "grad_norm": 0.0018028619233518839, + "learning_rate": 0.2456680522157359, + "loss": 0.1167, + "num_input_tokens_seen": 14520480, + "step": 11195 + }, + { + "epoch": 0.5472357266752986, + "grad_norm": 0.0009778968524187803, + "learning_rate": 0.245622675562684, + "loss": 0.0954, + "num_input_tokens_seen": 14527072, + "step": 11200 + }, + { + "epoch": 0.5472357266752986, + "eval_loss": 0.11581385135650635, + "eval_runtime": 402.7735, + "eval_samples_per_second": 90.336, + "eval_steps_per_second": 22.586, + "num_input_tokens_seen": 14527072, + "step": 11200 + }, + { + "epoch": 0.547480028338993, + "grad_norm": 0.0017773914150893688, + "learning_rate": 0.24557728416341384, + "loss": 0.1058, + "num_input_tokens_seen": 14533888, + "step": 11205 + }, + { + "epoch": 0.5477243300026873, + "grad_norm": 0.0012074729893356562, + "learning_rate": 0.24553187802492538, + "loss": 0.11, + "num_input_tokens_seen": 14540224, + "step": 11210 + }, + { + "epoch": 0.5479686316663817, + "grad_norm": 0.0018558134324848652, + "learning_rate": 0.24548645715422074, + "loss": 0.1309, + "num_input_tokens_seen": 14547264, + "step": 11215 + }, + { + "epoch": 0.5482129333300759, + "grad_norm": 0.0011238940060138702, + "learning_rate": 0.2454410215583045, + "loss": 0.0952, + "num_input_tokens_seen": 14553952, + "step": 11220 + }, + { + "epoch": 0.5484572349937703, + "grad_norm": 0.0017263019690290093, + "learning_rate": 0.24539557124418332, + "loss": 0.1054, + "num_input_tokens_seen": 14560128, + "step": 11225 + }, + { + "epoch": 0.5487015366574647, + "grad_norm": 0.00154199730604887, + "learning_rate": 0.24535010621886624, + "loss": 0.1406, + "num_input_tokens_seen": 14566368, + "step": 11230 + }, + { + "epoch": 0.548945838321159, + "grad_norm": 0.001285576494410634, + "learning_rate": 0.2453046264893646, + "loss": 0.0858, + "num_input_tokens_seen": 14572960, + "step": 11235 + }, + { + "epoch": 0.5491901399848533, + "grad_norm": 0.0017508980818092823, + "learning_rate": 0.24525913206269184, + "loss": 0.135, + "num_input_tokens_seen": 14579424, + "step": 11240 + }, + { + "epoch": 0.5494344416485476, + "grad_norm": 0.0032836163882166147, + "learning_rate": 0.2452136229458638, + "loss": 0.1322, + "num_input_tokens_seen": 14585696, + "step": 11245 + }, + { + "epoch": 0.549678743312242, + "grad_norm": 0.0015208754921332002, + "learning_rate": 0.24516809914589857, + "loss": 0.0919, + "num_input_tokens_seen": 14591968, + "step": 11250 + }, + { + "epoch": 0.5499230449759362, + "grad_norm": 0.0015789070166647434, + "learning_rate": 0.2451225606698165, + "loss": 0.1191, + "num_input_tokens_seen": 14598144, + "step": 11255 + }, + { + "epoch": 0.5501673466396306, + "grad_norm": 0.002115763258188963, + "learning_rate": 0.2450770075246402, + "loss": 0.0799, + "num_input_tokens_seen": 14605024, + "step": 11260 + }, + { + "epoch": 0.5504116483033249, + "grad_norm": 0.0016656288644298911, + "learning_rate": 0.24503143971739455, + "loss": 0.0779, + "num_input_tokens_seen": 14611232, + "step": 11265 + }, + { + "epoch": 0.5506559499670193, + "grad_norm": 0.001983663300052285, + "learning_rate": 0.24498585725510663, + "loss": 0.0825, + "num_input_tokens_seen": 14618048, + "step": 11270 + }, + { + "epoch": 0.5509002516307137, + "grad_norm": 0.0014986100140959024, + "learning_rate": 0.24494026014480583, + "loss": 0.1273, + "num_input_tokens_seen": 14624416, + "step": 11275 + }, + { + "epoch": 0.5511445532944079, + "grad_norm": 0.0016263348516076803, + "learning_rate": 0.24489464839352387, + "loss": 0.1227, + "num_input_tokens_seen": 14630432, + "step": 11280 + }, + { + "epoch": 0.5513888549581023, + "grad_norm": 0.0020681945607066154, + "learning_rate": 0.2448490220082946, + "loss": 0.1323, + "num_input_tokens_seen": 14637696, + "step": 11285 + }, + { + "epoch": 0.5516331566217966, + "grad_norm": 0.0008560725254938006, + "learning_rate": 0.24480338099615415, + "loss": 0.102, + "num_input_tokens_seen": 14644224, + "step": 11290 + }, + { + "epoch": 0.551877458285491, + "grad_norm": 0.0015458551933988929, + "learning_rate": 0.244757725364141, + "loss": 0.1202, + "num_input_tokens_seen": 14650560, + "step": 11295 + }, + { + "epoch": 0.5521217599491852, + "grad_norm": 0.0008292314596474171, + "learning_rate": 0.24471205511929583, + "loss": 0.1083, + "num_input_tokens_seen": 14657312, + "step": 11300 + }, + { + "epoch": 0.5523660616128796, + "grad_norm": 0.0014661946333944798, + "learning_rate": 0.24466637026866145, + "loss": 0.1255, + "num_input_tokens_seen": 14663520, + "step": 11305 + }, + { + "epoch": 0.5526103632765739, + "grad_norm": 0.001428675022907555, + "learning_rate": 0.2446206708192832, + "loss": 0.0875, + "num_input_tokens_seen": 14670304, + "step": 11310 + }, + { + "epoch": 0.5528546649402682, + "grad_norm": 0.0026945997960865498, + "learning_rate": 0.2445749567782084, + "loss": 0.0957, + "num_input_tokens_seen": 14676480, + "step": 11315 + }, + { + "epoch": 0.5530989666039626, + "grad_norm": 0.0012068471405655146, + "learning_rate": 0.2445292281524868, + "loss": 0.1064, + "num_input_tokens_seen": 14682944, + "step": 11320 + }, + { + "epoch": 0.5533432682676569, + "grad_norm": 0.001895078574307263, + "learning_rate": 0.24448348494917022, + "loss": 0.1026, + "num_input_tokens_seen": 14689568, + "step": 11325 + }, + { + "epoch": 0.5535875699313513, + "grad_norm": 0.0016049318946897984, + "learning_rate": 0.24443772717531295, + "loss": 0.114, + "num_input_tokens_seen": 14696064, + "step": 11330 + }, + { + "epoch": 0.5538318715950455, + "grad_norm": 0.0009523323969915509, + "learning_rate": 0.24439195483797138, + "loss": 0.0798, + "num_input_tokens_seen": 14702464, + "step": 11335 + }, + { + "epoch": 0.5540761732587399, + "grad_norm": 0.0012892765225842595, + "learning_rate": 0.24434616794420416, + "loss": 0.1311, + "num_input_tokens_seen": 14709184, + "step": 11340 + }, + { + "epoch": 0.5543204749224342, + "grad_norm": 0.0019412097753956914, + "learning_rate": 0.24430036650107223, + "loss": 0.1153, + "num_input_tokens_seen": 14716096, + "step": 11345 + }, + { + "epoch": 0.5545647765861286, + "grad_norm": 0.0018839450785890222, + "learning_rate": 0.2442545505156387, + "loss": 0.1089, + "num_input_tokens_seen": 14722752, + "step": 11350 + }, + { + "epoch": 0.5548090782498228, + "grad_norm": 0.0015076635172590613, + "learning_rate": 0.24420871999496904, + "loss": 0.0876, + "num_input_tokens_seen": 14729216, + "step": 11355 + }, + { + "epoch": 0.5550533799135172, + "grad_norm": 0.003294466994702816, + "learning_rate": 0.24416287494613084, + "loss": 0.1219, + "num_input_tokens_seen": 14735744, + "step": 11360 + }, + { + "epoch": 0.5552976815772115, + "grad_norm": 0.002478353213518858, + "learning_rate": 0.24411701537619399, + "loss": 0.1346, + "num_input_tokens_seen": 14742112, + "step": 11365 + }, + { + "epoch": 0.5555419832409059, + "grad_norm": 0.0015347194857895374, + "learning_rate": 0.24407114129223062, + "loss": 0.1121, + "num_input_tokens_seen": 14748192, + "step": 11370 + }, + { + "epoch": 0.5557862849046002, + "grad_norm": 0.0020096369553357363, + "learning_rate": 0.2440252527013151, + "loss": 0.1414, + "num_input_tokens_seen": 14754432, + "step": 11375 + }, + { + "epoch": 0.5560305865682945, + "grad_norm": 0.0016868861857801676, + "learning_rate": 0.24397934961052403, + "loss": 0.1174, + "num_input_tokens_seen": 14760896, + "step": 11380 + }, + { + "epoch": 0.5562748882319889, + "grad_norm": 0.0012782831909134984, + "learning_rate": 0.24393343202693618, + "loss": 0.1128, + "num_input_tokens_seen": 14767680, + "step": 11385 + }, + { + "epoch": 0.5565191898956832, + "grad_norm": 0.002434576628729701, + "learning_rate": 0.2438874999576327, + "loss": 0.1055, + "num_input_tokens_seen": 14773824, + "step": 11390 + }, + { + "epoch": 0.5567634915593775, + "grad_norm": 0.0017287294613197446, + "learning_rate": 0.24384155340969688, + "loss": 0.1101, + "num_input_tokens_seen": 14780544, + "step": 11395 + }, + { + "epoch": 0.5570077932230718, + "grad_norm": 0.0024389575701206923, + "learning_rate": 0.24379559239021423, + "loss": 0.1284, + "num_input_tokens_seen": 14787040, + "step": 11400 + }, + { + "epoch": 0.5570077932230718, + "eval_loss": 0.11395267397165298, + "eval_runtime": 402.7503, + "eval_samples_per_second": 90.341, + "eval_steps_per_second": 22.587, + "num_input_tokens_seen": 14787040, + "step": 11400 + }, + { + "epoch": 0.5572520948867662, + "grad_norm": 0.0011832150630652905, + "learning_rate": 0.2437496169062725, + "loss": 0.1228, + "num_input_tokens_seen": 14793568, + "step": 11405 + }, + { + "epoch": 0.5574963965504605, + "grad_norm": 0.002184587763622403, + "learning_rate": 0.24370362696496176, + "loss": 0.1518, + "num_input_tokens_seen": 14800096, + "step": 11410 + }, + { + "epoch": 0.5577406982141548, + "grad_norm": 0.0013009199174121022, + "learning_rate": 0.24365762257337417, + "loss": 0.1373, + "num_input_tokens_seen": 14806304, + "step": 11415 + }, + { + "epoch": 0.5579849998778492, + "grad_norm": 0.0012713023461401463, + "learning_rate": 0.2436116037386042, + "loss": 0.1042, + "num_input_tokens_seen": 14812576, + "step": 11420 + }, + { + "epoch": 0.5582293015415435, + "grad_norm": 0.0015254468889907002, + "learning_rate": 0.24356557046774852, + "loss": 0.1235, + "num_input_tokens_seen": 14819264, + "step": 11425 + }, + { + "epoch": 0.5584736032052379, + "grad_norm": 0.0019373849499970675, + "learning_rate": 0.24351952276790606, + "loss": 0.1104, + "num_input_tokens_seen": 14825888, + "step": 11430 + }, + { + "epoch": 0.5587179048689321, + "grad_norm": 0.0022877042647451162, + "learning_rate": 0.24347346064617797, + "loss": 0.1004, + "num_input_tokens_seen": 14832864, + "step": 11435 + }, + { + "epoch": 0.5589622065326265, + "grad_norm": 0.002225275384262204, + "learning_rate": 0.24342738410966758, + "loss": 0.1112, + "num_input_tokens_seen": 14839392, + "step": 11440 + }, + { + "epoch": 0.5592065081963208, + "grad_norm": 0.0033781242091208696, + "learning_rate": 0.24338129316548046, + "loss": 0.1769, + "num_input_tokens_seen": 14845952, + "step": 11445 + }, + { + "epoch": 0.5594508098600152, + "grad_norm": 0.002704952610656619, + "learning_rate": 0.24333518782072444, + "loss": 0.1276, + "num_input_tokens_seen": 14852832, + "step": 11450 + }, + { + "epoch": 0.5596951115237094, + "grad_norm": 0.0016089119017124176, + "learning_rate": 0.24328906808250952, + "loss": 0.1464, + "num_input_tokens_seen": 14858976, + "step": 11455 + }, + { + "epoch": 0.5599394131874038, + "grad_norm": 0.0011421851813793182, + "learning_rate": 0.243242933957948, + "loss": 0.111, + "num_input_tokens_seen": 14865696, + "step": 11460 + }, + { + "epoch": 0.5601837148510982, + "grad_norm": 0.0013675530208274722, + "learning_rate": 0.24319678545415427, + "loss": 0.13, + "num_input_tokens_seen": 14872192, + "step": 11465 + }, + { + "epoch": 0.5604280165147925, + "grad_norm": 0.0025732165668159723, + "learning_rate": 0.24315062257824507, + "loss": 0.1216, + "num_input_tokens_seen": 14878464, + "step": 11470 + }, + { + "epoch": 0.5606723181784868, + "grad_norm": 0.001260278164409101, + "learning_rate": 0.24310444533733921, + "loss": 0.1267, + "num_input_tokens_seen": 14884672, + "step": 11475 + }, + { + "epoch": 0.5609166198421811, + "grad_norm": 0.0009347221930511296, + "learning_rate": 0.2430582537385579, + "loss": 0.0969, + "num_input_tokens_seen": 14891136, + "step": 11480 + }, + { + "epoch": 0.5611609215058755, + "grad_norm": 0.0015056222910061479, + "learning_rate": 0.2430120477890244, + "loss": 0.104, + "num_input_tokens_seen": 14897312, + "step": 11485 + }, + { + "epoch": 0.5614052231695698, + "grad_norm": 0.001937590423040092, + "learning_rate": 0.24296582749586426, + "loss": 0.1122, + "num_input_tokens_seen": 14903360, + "step": 11490 + }, + { + "epoch": 0.5616495248332641, + "grad_norm": 0.0010741751175373793, + "learning_rate": 0.24291959286620526, + "loss": 0.125, + "num_input_tokens_seen": 14909824, + "step": 11495 + }, + { + "epoch": 0.5618938264969584, + "grad_norm": 0.0017459038645029068, + "learning_rate": 0.24287334390717738, + "loss": 0.1258, + "num_input_tokens_seen": 14916128, + "step": 11500 + }, + { + "epoch": 0.5621381281606528, + "grad_norm": 0.0014132227515801787, + "learning_rate": 0.24282708062591268, + "loss": 0.108, + "num_input_tokens_seen": 14922656, + "step": 11505 + }, + { + "epoch": 0.562382429824347, + "grad_norm": 0.001221425598487258, + "learning_rate": 0.24278080302954563, + "loss": 0.1164, + "num_input_tokens_seen": 14928896, + "step": 11510 + }, + { + "epoch": 0.5626267314880414, + "grad_norm": 0.0012052204692736268, + "learning_rate": 0.24273451112521283, + "loss": 0.0774, + "num_input_tokens_seen": 14935328, + "step": 11515 + }, + { + "epoch": 0.5628710331517358, + "grad_norm": 0.0015065264888107777, + "learning_rate": 0.242688204920053, + "loss": 0.1255, + "num_input_tokens_seen": 14942112, + "step": 11520 + }, + { + "epoch": 0.5631153348154301, + "grad_norm": 0.0010948526905849576, + "learning_rate": 0.24264188442120715, + "loss": 0.0907, + "num_input_tokens_seen": 14949216, + "step": 11525 + }, + { + "epoch": 0.5633596364791245, + "grad_norm": 0.0010763512691482902, + "learning_rate": 0.24259554963581853, + "loss": 0.0834, + "num_input_tokens_seen": 14955392, + "step": 11530 + }, + { + "epoch": 0.5636039381428187, + "grad_norm": 0.001158498926088214, + "learning_rate": 0.24254920057103257, + "loss": 0.1112, + "num_input_tokens_seen": 14961856, + "step": 11535 + }, + { + "epoch": 0.5638482398065131, + "grad_norm": 0.0011312643764540553, + "learning_rate": 0.24250283723399685, + "loss": 0.1009, + "num_input_tokens_seen": 14968448, + "step": 11540 + }, + { + "epoch": 0.5640925414702074, + "grad_norm": 0.0018821106059476733, + "learning_rate": 0.24245645963186108, + "loss": 0.1205, + "num_input_tokens_seen": 14975040, + "step": 11545 + }, + { + "epoch": 0.5643368431339018, + "grad_norm": 0.0024614848662167788, + "learning_rate": 0.2424100677717774, + "loss": 0.1139, + "num_input_tokens_seen": 14981344, + "step": 11550 + }, + { + "epoch": 0.564581144797596, + "grad_norm": 0.0007948394049890339, + "learning_rate": 0.24236366166090004, + "loss": 0.0971, + "num_input_tokens_seen": 14987552, + "step": 11555 + }, + { + "epoch": 0.5648254464612904, + "grad_norm": 0.0015990306856110692, + "learning_rate": 0.24231724130638527, + "loss": 0.1099, + "num_input_tokens_seen": 14993824, + "step": 11560 + }, + { + "epoch": 0.5650697481249848, + "grad_norm": 0.0014295704895630479, + "learning_rate": 0.2422708067153917, + "loss": 0.1247, + "num_input_tokens_seen": 15000384, + "step": 11565 + }, + { + "epoch": 0.565314049788679, + "grad_norm": 0.0010567022254690528, + "learning_rate": 0.24222435789508026, + "loss": 0.1183, + "num_input_tokens_seen": 15006848, + "step": 11570 + }, + { + "epoch": 0.5655583514523734, + "grad_norm": 0.0010339751606807113, + "learning_rate": 0.24217789485261387, + "loss": 0.1409, + "num_input_tokens_seen": 15013312, + "step": 11575 + }, + { + "epoch": 0.5658026531160677, + "grad_norm": 0.001597579219378531, + "learning_rate": 0.2421314175951577, + "loss": 0.1242, + "num_input_tokens_seen": 15019488, + "step": 11580 + }, + { + "epoch": 0.5660469547797621, + "grad_norm": 0.001557440496981144, + "learning_rate": 0.2420849261298791, + "loss": 0.1378, + "num_input_tokens_seen": 15025888, + "step": 11585 + }, + { + "epoch": 0.5662912564434563, + "grad_norm": 0.0015267248963937163, + "learning_rate": 0.24203842046394775, + "loss": 0.1094, + "num_input_tokens_seen": 15032768, + "step": 11590 + }, + { + "epoch": 0.5665355581071507, + "grad_norm": 0.0012487717904150486, + "learning_rate": 0.24199190060453535, + "loss": 0.1095, + "num_input_tokens_seen": 15039424, + "step": 11595 + }, + { + "epoch": 0.566779859770845, + "grad_norm": 0.0014452928444370627, + "learning_rate": 0.2419453665588158, + "loss": 0.1099, + "num_input_tokens_seen": 15045600, + "step": 11600 + }, + { + "epoch": 0.566779859770845, + "eval_loss": 0.11164794117212296, + "eval_runtime": 402.5499, + "eval_samples_per_second": 90.386, + "eval_steps_per_second": 22.598, + "num_input_tokens_seen": 15045600, + "step": 11600 + }, + { + "epoch": 0.5670241614345394, + "grad_norm": 0.001318547292612493, + "learning_rate": 0.24189881833396523, + "loss": 0.1121, + "num_input_tokens_seen": 15051872, + "step": 11605 + }, + { + "epoch": 0.5672684630982336, + "grad_norm": 0.0015450221253558993, + "learning_rate": 0.24185225593716203, + "loss": 0.1104, + "num_input_tokens_seen": 15058400, + "step": 11610 + }, + { + "epoch": 0.567512764761928, + "grad_norm": 0.0021991138346493244, + "learning_rate": 0.2418056793755867, + "loss": 0.09, + "num_input_tokens_seen": 15064896, + "step": 11615 + }, + { + "epoch": 0.5677570664256224, + "grad_norm": 0.0026518201921135187, + "learning_rate": 0.24175908865642187, + "loss": 0.1123, + "num_input_tokens_seen": 15071584, + "step": 11620 + }, + { + "epoch": 0.5680013680893167, + "grad_norm": 0.0021245419047772884, + "learning_rate": 0.24171248378685248, + "loss": 0.1211, + "num_input_tokens_seen": 15077824, + "step": 11625 + }, + { + "epoch": 0.568245669753011, + "grad_norm": 0.0014747640816494823, + "learning_rate": 0.24166586477406554, + "loss": 0.0822, + "num_input_tokens_seen": 15084160, + "step": 11630 + }, + { + "epoch": 0.5684899714167053, + "grad_norm": 0.0027814176864922047, + "learning_rate": 0.24161923162525034, + "loss": 0.0937, + "num_input_tokens_seen": 15090880, + "step": 11635 + }, + { + "epoch": 0.5687342730803997, + "grad_norm": 0.0015725069679319859, + "learning_rate": 0.2415725843475982, + "loss": 0.1228, + "num_input_tokens_seen": 15097184, + "step": 11640 + }, + { + "epoch": 0.568978574744094, + "grad_norm": 0.0014006971614435315, + "learning_rate": 0.24152592294830286, + "loss": 0.1284, + "num_input_tokens_seen": 15103552, + "step": 11645 + }, + { + "epoch": 0.5692228764077883, + "grad_norm": 0.0013920129276812077, + "learning_rate": 0.24147924743455995, + "loss": 0.0925, + "num_input_tokens_seen": 15110176, + "step": 11650 + }, + { + "epoch": 0.5694671780714826, + "grad_norm": 0.002987067447975278, + "learning_rate": 0.24143255781356754, + "loss": 0.1262, + "num_input_tokens_seen": 15116736, + "step": 11655 + }, + { + "epoch": 0.569711479735177, + "grad_norm": 0.0013322869781404734, + "learning_rate": 0.24138585409252566, + "loss": 0.1162, + "num_input_tokens_seen": 15123936, + "step": 11660 + }, + { + "epoch": 0.5699557813988714, + "grad_norm": 0.0011435127817094326, + "learning_rate": 0.24133913627863662, + "loss": 0.0906, + "num_input_tokens_seen": 15130368, + "step": 11665 + }, + { + "epoch": 0.5702000830625656, + "grad_norm": 0.0019906319212168455, + "learning_rate": 0.241292404379105, + "loss": 0.1258, + "num_input_tokens_seen": 15136928, + "step": 11670 + }, + { + "epoch": 0.57044438472626, + "grad_norm": 0.0023459424264729023, + "learning_rate": 0.24124565840113735, + "loss": 0.1006, + "num_input_tokens_seen": 15144032, + "step": 11675 + }, + { + "epoch": 0.5706886863899543, + "grad_norm": 0.001703434158116579, + "learning_rate": 0.2411988983519425, + "loss": 0.0955, + "num_input_tokens_seen": 15150528, + "step": 11680 + }, + { + "epoch": 0.5709329880536487, + "grad_norm": 0.0013600512174889445, + "learning_rate": 0.24115212423873145, + "loss": 0.0982, + "num_input_tokens_seen": 15156800, + "step": 11685 + }, + { + "epoch": 0.5711772897173429, + "grad_norm": 0.001391325960867107, + "learning_rate": 0.24110533606871737, + "loss": 0.1156, + "num_input_tokens_seen": 15163072, + "step": 11690 + }, + { + "epoch": 0.5714215913810373, + "grad_norm": 0.0010905131930485368, + "learning_rate": 0.24105853384911552, + "loss": 0.0905, + "num_input_tokens_seen": 15169568, + "step": 11695 + }, + { + "epoch": 0.5716658930447316, + "grad_norm": 0.0015948658110573888, + "learning_rate": 0.24101171758714346, + "loss": 0.1202, + "num_input_tokens_seen": 15175904, + "step": 11700 + }, + { + "epoch": 0.571910194708426, + "grad_norm": 0.0017204697942361236, + "learning_rate": 0.24096488729002086, + "loss": 0.1378, + "num_input_tokens_seen": 15181728, + "step": 11705 + }, + { + "epoch": 0.5721544963721203, + "grad_norm": 0.001709102070890367, + "learning_rate": 0.24091804296496946, + "loss": 0.124, + "num_input_tokens_seen": 15188288, + "step": 11710 + }, + { + "epoch": 0.5723987980358146, + "grad_norm": 0.001620293129235506, + "learning_rate": 0.2408711846192133, + "loss": 0.1055, + "num_input_tokens_seen": 15194848, + "step": 11715 + }, + { + "epoch": 0.572643099699509, + "grad_norm": 0.0017812387086451054, + "learning_rate": 0.24082431225997855, + "loss": 0.1285, + "num_input_tokens_seen": 15201536, + "step": 11720 + }, + { + "epoch": 0.5728874013632033, + "grad_norm": 0.001677782041952014, + "learning_rate": 0.24077742589449344, + "loss": 0.1281, + "num_input_tokens_seen": 15208032, + "step": 11725 + }, + { + "epoch": 0.5731317030268976, + "grad_norm": 0.0013796011917293072, + "learning_rate": 0.24073052552998844, + "loss": 0.1298, + "num_input_tokens_seen": 15214272, + "step": 11730 + }, + { + "epoch": 0.5733760046905919, + "grad_norm": 0.0009304159320890903, + "learning_rate": 0.2406836111736963, + "loss": 0.104, + "num_input_tokens_seen": 15220672, + "step": 11735 + }, + { + "epoch": 0.5736203063542863, + "grad_norm": 0.0012493119575083256, + "learning_rate": 0.2406366828328517, + "loss": 0.1075, + "num_input_tokens_seen": 15227136, + "step": 11740 + }, + { + "epoch": 0.5738646080179806, + "grad_norm": 0.0009395433007739484, + "learning_rate": 0.2405897405146915, + "loss": 0.1239, + "num_input_tokens_seen": 15233536, + "step": 11745 + }, + { + "epoch": 0.5741089096816749, + "grad_norm": 0.0017960087861865759, + "learning_rate": 0.240542784226455, + "loss": 0.1217, + "num_input_tokens_seen": 15240512, + "step": 11750 + }, + { + "epoch": 0.5743532113453692, + "grad_norm": 0.0013072504661977291, + "learning_rate": 0.24049581397538328, + "loss": 0.0998, + "num_input_tokens_seen": 15246784, + "step": 11755 + }, + { + "epoch": 0.5745975130090636, + "grad_norm": 0.0028275400400161743, + "learning_rate": 0.24044882976871984, + "loss": 0.1219, + "num_input_tokens_seen": 15252896, + "step": 11760 + }, + { + "epoch": 0.574841814672758, + "grad_norm": 0.0017289331881329417, + "learning_rate": 0.2404018316137102, + "loss": 0.118, + "num_input_tokens_seen": 15259232, + "step": 11765 + }, + { + "epoch": 0.5750861163364522, + "grad_norm": 0.0018260376527905464, + "learning_rate": 0.24035481951760204, + "loss": 0.0859, + "num_input_tokens_seen": 15265760, + "step": 11770 + }, + { + "epoch": 0.5753304180001466, + "grad_norm": 0.0032464771065860987, + "learning_rate": 0.2403077934876452, + "loss": 0.1265, + "num_input_tokens_seen": 15272416, + "step": 11775 + }, + { + "epoch": 0.5755747196638409, + "grad_norm": 0.0017124894075095654, + "learning_rate": 0.2402607535310918, + "loss": 0.082, + "num_input_tokens_seen": 15278624, + "step": 11780 + }, + { + "epoch": 0.5758190213275353, + "grad_norm": 0.0011474089697003365, + "learning_rate": 0.2402136996551959, + "loss": 0.115, + "num_input_tokens_seen": 15284960, + "step": 11785 + }, + { + "epoch": 0.5760633229912295, + "grad_norm": 0.0019029222894459963, + "learning_rate": 0.24016663186721376, + "loss": 0.119, + "num_input_tokens_seen": 15292960, + "step": 11790 + }, + { + "epoch": 0.5763076246549239, + "grad_norm": 0.002032611519098282, + "learning_rate": 0.24011955017440395, + "loss": 0.1134, + "num_input_tokens_seen": 15299648, + "step": 11795 + }, + { + "epoch": 0.5765519263186182, + "grad_norm": 0.0015277423663064837, + "learning_rate": 0.24007245458402696, + "loss": 0.0898, + "num_input_tokens_seen": 15306176, + "step": 11800 + }, + { + "epoch": 0.5765519263186182, + "eval_loss": 0.1122361421585083, + "eval_runtime": 401.9577, + "eval_samples_per_second": 90.519, + "eval_steps_per_second": 22.632, + "num_input_tokens_seen": 15306176, + "step": 11800 + }, + { + "epoch": 0.5767962279823126, + "grad_norm": 0.0011827271664515138, + "learning_rate": 0.2400253451033456, + "loss": 0.1469, + "num_input_tokens_seen": 15312416, + "step": 11805 + }, + { + "epoch": 0.5770405296460069, + "grad_norm": 0.0010047375690191984, + "learning_rate": 0.23997822173962463, + "loss": 0.0968, + "num_input_tokens_seen": 15318656, + "step": 11810 + }, + { + "epoch": 0.5772848313097012, + "grad_norm": 0.00103290774859488, + "learning_rate": 0.23993108450013118, + "loss": 0.089, + "num_input_tokens_seen": 15325376, + "step": 11815 + }, + { + "epoch": 0.5775291329733956, + "grad_norm": 0.0014592770021408796, + "learning_rate": 0.2398839333921343, + "loss": 0.1221, + "num_input_tokens_seen": 15331872, + "step": 11820 + }, + { + "epoch": 0.5777734346370899, + "grad_norm": 0.0012722001411020756, + "learning_rate": 0.23983676842290536, + "loss": 0.0987, + "num_input_tokens_seen": 15338240, + "step": 11825 + }, + { + "epoch": 0.5780177363007842, + "grad_norm": 0.0011032192269340158, + "learning_rate": 0.2397895895997178, + "loss": 0.0734, + "num_input_tokens_seen": 15345184, + "step": 11830 + }, + { + "epoch": 0.5782620379644785, + "grad_norm": 0.0017443806864321232, + "learning_rate": 0.23974239692984714, + "loss": 0.1204, + "num_input_tokens_seen": 15351744, + "step": 11835 + }, + { + "epoch": 0.5785063396281729, + "grad_norm": 0.001426875707693398, + "learning_rate": 0.2396951904205711, + "loss": 0.1203, + "num_input_tokens_seen": 15358048, + "step": 11840 + }, + { + "epoch": 0.5787506412918672, + "grad_norm": 0.0013794427504763007, + "learning_rate": 0.23964797007916952, + "loss": 0.1073, + "num_input_tokens_seen": 15364288, + "step": 11845 + }, + { + "epoch": 0.5789949429555615, + "grad_norm": 0.0009837074903771281, + "learning_rate": 0.23960073591292436, + "loss": 0.1353, + "num_input_tokens_seen": 15370848, + "step": 11850 + }, + { + "epoch": 0.5792392446192559, + "grad_norm": 0.001238429918885231, + "learning_rate": 0.2395534879291197, + "loss": 0.1154, + "num_input_tokens_seen": 15377088, + "step": 11855 + }, + { + "epoch": 0.5794835462829502, + "grad_norm": 0.0011017810320481658, + "learning_rate": 0.23950622613504186, + "loss": 0.1171, + "num_input_tokens_seen": 15383808, + "step": 11860 + }, + { + "epoch": 0.5797278479466446, + "grad_norm": 0.0010625721188262105, + "learning_rate": 0.2394589505379791, + "loss": 0.108, + "num_input_tokens_seen": 15390624, + "step": 11865 + }, + { + "epoch": 0.5799721496103388, + "grad_norm": 0.0011119891423732042, + "learning_rate": 0.23941166114522197, + "loss": 0.1312, + "num_input_tokens_seen": 15397280, + "step": 11870 + }, + { + "epoch": 0.5802164512740332, + "grad_norm": 0.001379081280902028, + "learning_rate": 0.23936435796406308, + "loss": 0.0953, + "num_input_tokens_seen": 15404096, + "step": 11875 + }, + { + "epoch": 0.5804607529377275, + "grad_norm": 0.0015813177451491356, + "learning_rate": 0.23931704100179715, + "loss": 0.1182, + "num_input_tokens_seen": 15410464, + "step": 11880 + }, + { + "epoch": 0.5807050546014219, + "grad_norm": 0.0011183293536305428, + "learning_rate": 0.2392697102657211, + "loss": 0.1212, + "num_input_tokens_seen": 15416736, + "step": 11885 + }, + { + "epoch": 0.5809493562651161, + "grad_norm": 0.0012102625332772732, + "learning_rate": 0.23922236576313388, + "loss": 0.1047, + "num_input_tokens_seen": 15422720, + "step": 11890 + }, + { + "epoch": 0.5811936579288105, + "grad_norm": 0.0010288865305483341, + "learning_rate": 0.2391750075013366, + "loss": 0.1058, + "num_input_tokens_seen": 15429184, + "step": 11895 + }, + { + "epoch": 0.5814379595925048, + "grad_norm": 0.0012024604948237538, + "learning_rate": 0.2391276354876326, + "loss": 0.1066, + "num_input_tokens_seen": 15435584, + "step": 11900 + }, + { + "epoch": 0.5816822612561992, + "grad_norm": 0.001683708280324936, + "learning_rate": 0.23908024972932707, + "loss": 0.0782, + "num_input_tokens_seen": 15442176, + "step": 11905 + }, + { + "epoch": 0.5819265629198935, + "grad_norm": 0.0013687602477148175, + "learning_rate": 0.2390328502337276, + "loss": 0.1566, + "num_input_tokens_seen": 15448384, + "step": 11910 + }, + { + "epoch": 0.5821708645835878, + "grad_norm": 0.001590246451087296, + "learning_rate": 0.23898543700814376, + "loss": 0.1009, + "num_input_tokens_seen": 15454848, + "step": 11915 + }, + { + "epoch": 0.5824151662472822, + "grad_norm": 0.0010712285293266177, + "learning_rate": 0.2389380100598873, + "loss": 0.0983, + "num_input_tokens_seen": 15461376, + "step": 11920 + }, + { + "epoch": 0.5826594679109764, + "grad_norm": 0.0015578127931803465, + "learning_rate": 0.23889056939627207, + "loss": 0.1081, + "num_input_tokens_seen": 15468448, + "step": 11925 + }, + { + "epoch": 0.5829037695746708, + "grad_norm": 0.001238176366314292, + "learning_rate": 0.23884311502461386, + "loss": 0.1381, + "num_input_tokens_seen": 15474880, + "step": 11930 + }, + { + "epoch": 0.5831480712383651, + "grad_norm": 0.002311030635610223, + "learning_rate": 0.23879564695223088, + "loss": 0.1088, + "num_input_tokens_seen": 15480928, + "step": 11935 + }, + { + "epoch": 0.5833923729020595, + "grad_norm": 0.0018359377281740308, + "learning_rate": 0.23874816518644332, + "loss": 0.1058, + "num_input_tokens_seen": 15486976, + "step": 11940 + }, + { + "epoch": 0.5836366745657537, + "grad_norm": 0.0015617316821590066, + "learning_rate": 0.23870066973457335, + "loss": 0.1129, + "num_input_tokens_seen": 15493696, + "step": 11945 + }, + { + "epoch": 0.5838809762294481, + "grad_norm": 0.002083169063553214, + "learning_rate": 0.23865316060394545, + "loss": 0.1148, + "num_input_tokens_seen": 15500352, + "step": 11950 + }, + { + "epoch": 0.5841252778931425, + "grad_norm": 0.001053516403771937, + "learning_rate": 0.2386056378018861, + "loss": 0.1139, + "num_input_tokens_seen": 15506304, + "step": 11955 + }, + { + "epoch": 0.5843695795568368, + "grad_norm": 0.0015384553698822856, + "learning_rate": 0.2385581013357239, + "loss": 0.1239, + "num_input_tokens_seen": 15512512, + "step": 11960 + }, + { + "epoch": 0.5846138812205312, + "grad_norm": 0.0012236356269568205, + "learning_rate": 0.23851055121278958, + "loss": 0.1259, + "num_input_tokens_seen": 15519136, + "step": 11965 + }, + { + "epoch": 0.5848581828842254, + "grad_norm": 0.0014309228863567114, + "learning_rate": 0.23846298744041594, + "loss": 0.1083, + "num_input_tokens_seen": 15525664, + "step": 11970 + }, + { + "epoch": 0.5851024845479198, + "grad_norm": 0.0011318294564262033, + "learning_rate": 0.23841541002593802, + "loss": 0.1098, + "num_input_tokens_seen": 15532064, + "step": 11975 + }, + { + "epoch": 0.5853467862116141, + "grad_norm": 0.0007187177543528378, + "learning_rate": 0.23836781897669276, + "loss": 0.0975, + "num_input_tokens_seen": 15538688, + "step": 11980 + }, + { + "epoch": 0.5855910878753084, + "grad_norm": 0.0012476478004828095, + "learning_rate": 0.23832021430001926, + "loss": 0.1188, + "num_input_tokens_seen": 15545344, + "step": 11985 + }, + { + "epoch": 0.5858353895390027, + "grad_norm": 0.0017969021573662758, + "learning_rate": 0.2382725960032588, + "loss": 0.1064, + "num_input_tokens_seen": 15551904, + "step": 11990 + }, + { + "epoch": 0.5860796912026971, + "grad_norm": 0.001153708784841001, + "learning_rate": 0.23822496409375482, + "loss": 0.108, + "num_input_tokens_seen": 15558400, + "step": 11995 + }, + { + "epoch": 0.5863239928663915, + "grad_norm": 0.0010171188041567802, + "learning_rate": 0.2381773185788526, + "loss": 0.1014, + "num_input_tokens_seen": 15565184, + "step": 12000 + }, + { + "epoch": 0.5863239928663915, + "eval_loss": 0.11005989462137222, + "eval_runtime": 402.3126, + "eval_samples_per_second": 90.44, + "eval_steps_per_second": 22.612, + "num_input_tokens_seen": 15565184, + "step": 12000 + }, + { + "epoch": 0.5865682945300857, + "grad_norm": 0.0015140643808990717, + "learning_rate": 0.2381296594658998, + "loss": 0.0926, + "num_input_tokens_seen": 15572096, + "step": 12005 + }, + { + "epoch": 0.5868125961937801, + "grad_norm": 0.0010640190448611975, + "learning_rate": 0.238081986762246, + "loss": 0.1017, + "num_input_tokens_seen": 15578688, + "step": 12010 + }, + { + "epoch": 0.5870568978574744, + "grad_norm": 0.0014085060684010386, + "learning_rate": 0.23803430047524293, + "loss": 0.0941, + "num_input_tokens_seen": 15584896, + "step": 12015 + }, + { + "epoch": 0.5873011995211688, + "grad_norm": 0.007793162949383259, + "learning_rate": 0.23798660061224441, + "loss": 0.1301, + "num_input_tokens_seen": 15591392, + "step": 12020 + }, + { + "epoch": 0.587545501184863, + "grad_norm": 0.0011614178074523807, + "learning_rate": 0.23793888718060632, + "loss": 0.097, + "num_input_tokens_seen": 15597792, + "step": 12025 + }, + { + "epoch": 0.5877898028485574, + "grad_norm": 0.0016030187252908945, + "learning_rate": 0.23789116018768675, + "loss": 0.1269, + "num_input_tokens_seen": 15603968, + "step": 12030 + }, + { + "epoch": 0.5880341045122517, + "grad_norm": 0.0018213804578408599, + "learning_rate": 0.2378434196408458, + "loss": 0.1158, + "num_input_tokens_seen": 15610400, + "step": 12035 + }, + { + "epoch": 0.5882784061759461, + "grad_norm": 0.0016409712843596935, + "learning_rate": 0.23779566554744563, + "loss": 0.1116, + "num_input_tokens_seen": 15617056, + "step": 12040 + }, + { + "epoch": 0.5885227078396403, + "grad_norm": 0.0016002495540305972, + "learning_rate": 0.23774789791485051, + "loss": 0.0933, + "num_input_tokens_seen": 15624096, + "step": 12045 + }, + { + "epoch": 0.5887670095033347, + "grad_norm": 0.0016420213505625725, + "learning_rate": 0.2377001167504268, + "loss": 0.1298, + "num_input_tokens_seen": 15630208, + "step": 12050 + }, + { + "epoch": 0.5890113111670291, + "grad_norm": 0.0012045642361044884, + "learning_rate": 0.23765232206154302, + "loss": 0.0796, + "num_input_tokens_seen": 15637184, + "step": 12055 + }, + { + "epoch": 0.5892556128307234, + "grad_norm": 0.0011566239409148693, + "learning_rate": 0.23760451385556966, + "loss": 0.0903, + "num_input_tokens_seen": 15643552, + "step": 12060 + }, + { + "epoch": 0.5894999144944177, + "grad_norm": 0.0038011192809790373, + "learning_rate": 0.23755669213987932, + "loss": 0.118, + "num_input_tokens_seen": 15649984, + "step": 12065 + }, + { + "epoch": 0.589744216158112, + "grad_norm": 0.0015374976210296154, + "learning_rate": 0.23750885692184676, + "loss": 0.1005, + "num_input_tokens_seen": 15656448, + "step": 12070 + }, + { + "epoch": 0.5899885178218064, + "grad_norm": 0.0025869738310575485, + "learning_rate": 0.23746100820884875, + "loss": 0.0857, + "num_input_tokens_seen": 15662912, + "step": 12075 + }, + { + "epoch": 0.5902328194855007, + "grad_norm": 0.001597653841599822, + "learning_rate": 0.23741314600826421, + "loss": 0.1002, + "num_input_tokens_seen": 15669024, + "step": 12080 + }, + { + "epoch": 0.590477121149195, + "grad_norm": 0.0014349487610161304, + "learning_rate": 0.23736527032747406, + "loss": 0.1057, + "num_input_tokens_seen": 15675424, + "step": 12085 + }, + { + "epoch": 0.5907214228128893, + "grad_norm": 0.002687078434973955, + "learning_rate": 0.23731738117386128, + "loss": 0.1287, + "num_input_tokens_seen": 15681760, + "step": 12090 + }, + { + "epoch": 0.5909657244765837, + "grad_norm": 0.0014797502662986517, + "learning_rate": 0.237269478554811, + "loss": 0.1176, + "num_input_tokens_seen": 15688416, + "step": 12095 + }, + { + "epoch": 0.5912100261402781, + "grad_norm": 0.002034678589552641, + "learning_rate": 0.23722156247771053, + "loss": 0.1116, + "num_input_tokens_seen": 15694880, + "step": 12100 + }, + { + "epoch": 0.5914543278039723, + "grad_norm": 0.001043852069415152, + "learning_rate": 0.23717363294994895, + "loss": 0.121, + "num_input_tokens_seen": 15701760, + "step": 12105 + }, + { + "epoch": 0.5916986294676667, + "grad_norm": 0.0012600959744304419, + "learning_rate": 0.2371256899789177, + "loss": 0.1052, + "num_input_tokens_seen": 15708768, + "step": 12110 + }, + { + "epoch": 0.591942931131361, + "grad_norm": 0.0026835305616259575, + "learning_rate": 0.23707773357201017, + "loss": 0.1186, + "num_input_tokens_seen": 15715136, + "step": 12115 + }, + { + "epoch": 0.5921872327950554, + "grad_norm": 0.0010674366494640708, + "learning_rate": 0.2370297637366218, + "loss": 0.1074, + "num_input_tokens_seen": 15721760, + "step": 12120 + }, + { + "epoch": 0.5924315344587496, + "grad_norm": 0.0009911528322845697, + "learning_rate": 0.23698178048015026, + "loss": 0.1163, + "num_input_tokens_seen": 15727840, + "step": 12125 + }, + { + "epoch": 0.592675836122444, + "grad_norm": 0.0009286325075663626, + "learning_rate": 0.236933783809995, + "loss": 0.1036, + "num_input_tokens_seen": 15734240, + "step": 12130 + }, + { + "epoch": 0.5929201377861383, + "grad_norm": 0.0009774963837116957, + "learning_rate": 0.23688577373355785, + "loss": 0.1219, + "num_input_tokens_seen": 15740640, + "step": 12135 + }, + { + "epoch": 0.5931644394498327, + "grad_norm": 0.0014737161109223962, + "learning_rate": 0.23683775025824247, + "loss": 0.1009, + "num_input_tokens_seen": 15746848, + "step": 12140 + }, + { + "epoch": 0.593408741113527, + "grad_norm": 0.0020287686493247747, + "learning_rate": 0.2367897133914548, + "loss": 0.1552, + "num_input_tokens_seen": 15753344, + "step": 12145 + }, + { + "epoch": 0.5936530427772213, + "grad_norm": 0.0007548218709416687, + "learning_rate": 0.2367416631406026, + "loss": 0.1007, + "num_input_tokens_seen": 15759648, + "step": 12150 + }, + { + "epoch": 0.5938973444409157, + "grad_norm": 0.0014700426254421473, + "learning_rate": 0.23669359951309588, + "loss": 0.1039, + "num_input_tokens_seen": 15765760, + "step": 12155 + }, + { + "epoch": 0.59414164610461, + "grad_norm": 0.0016162481624633074, + "learning_rate": 0.23664552251634666, + "loss": 0.113, + "num_input_tokens_seen": 15772032, + "step": 12160 + }, + { + "epoch": 0.5943859477683043, + "grad_norm": 0.0011765433009713888, + "learning_rate": 0.23659743215776907, + "loss": 0.0893, + "num_input_tokens_seen": 15778400, + "step": 12165 + }, + { + "epoch": 0.5946302494319986, + "grad_norm": 0.0009115681750699878, + "learning_rate": 0.23654932844477908, + "loss": 0.0818, + "num_input_tokens_seen": 15785056, + "step": 12170 + }, + { + "epoch": 0.594874551095693, + "grad_norm": 0.0021609836257994175, + "learning_rate": 0.23650121138479507, + "loss": 0.1337, + "num_input_tokens_seen": 15791552, + "step": 12175 + }, + { + "epoch": 0.5951188527593873, + "grad_norm": 0.0016663889400660992, + "learning_rate": 0.23645308098523724, + "loss": 0.12, + "num_input_tokens_seen": 15797632, + "step": 12180 + }, + { + "epoch": 0.5953631544230816, + "grad_norm": 0.0013181301765143871, + "learning_rate": 0.23640493725352785, + "loss": 0.1301, + "num_input_tokens_seen": 15804448, + "step": 12185 + }, + { + "epoch": 0.5956074560867759, + "grad_norm": 0.0012329915771260858, + "learning_rate": 0.2363567801970913, + "loss": 0.0841, + "num_input_tokens_seen": 15811264, + "step": 12190 + }, + { + "epoch": 0.5958517577504703, + "grad_norm": 0.0013742762384936213, + "learning_rate": 0.236308609823354, + "loss": 0.1736, + "num_input_tokens_seen": 15817600, + "step": 12195 + }, + { + "epoch": 0.5960960594141647, + "grad_norm": 0.0012791340705007315, + "learning_rate": 0.23626042613974452, + "loss": 0.1279, + "num_input_tokens_seen": 15824576, + "step": 12200 + }, + { + "epoch": 0.5960960594141647, + "eval_loss": 0.12056899815797806, + "eval_runtime": 401.7121, + "eval_samples_per_second": 90.575, + "eval_steps_per_second": 22.646, + "num_input_tokens_seen": 15824576, + "step": 12200 + }, + { + "epoch": 0.5963403610778589, + "grad_norm": 0.001577141578309238, + "learning_rate": 0.23621222915369325, + "loss": 0.1129, + "num_input_tokens_seen": 15831360, + "step": 12205 + }, + { + "epoch": 0.5965846627415533, + "grad_norm": 0.0016324950847774744, + "learning_rate": 0.23616401887263283, + "loss": 0.1267, + "num_input_tokens_seen": 15837856, + "step": 12210 + }, + { + "epoch": 0.5968289644052476, + "grad_norm": 0.000990736298263073, + "learning_rate": 0.23611579530399793, + "loss": 0.1026, + "num_input_tokens_seen": 15844448, + "step": 12215 + }, + { + "epoch": 0.597073266068942, + "grad_norm": 0.0016432367265224457, + "learning_rate": 0.23606755845522517, + "loss": 0.109, + "num_input_tokens_seen": 15850976, + "step": 12220 + }, + { + "epoch": 0.5973175677326362, + "grad_norm": 0.0007243150030262768, + "learning_rate": 0.23601930833375329, + "loss": 0.0978, + "num_input_tokens_seen": 15857472, + "step": 12225 + }, + { + "epoch": 0.5975618693963306, + "grad_norm": 0.0016828509978950024, + "learning_rate": 0.23597104494702312, + "loss": 0.1174, + "num_input_tokens_seen": 15863744, + "step": 12230 + }, + { + "epoch": 0.5978061710600249, + "grad_norm": 0.0008966794121079147, + "learning_rate": 0.23592276830247744, + "loss": 0.1152, + "num_input_tokens_seen": 15870400, + "step": 12235 + }, + { + "epoch": 0.5980504727237193, + "grad_norm": 0.0018228372791782022, + "learning_rate": 0.2358744784075611, + "loss": 0.1131, + "num_input_tokens_seen": 15877280, + "step": 12240 + }, + { + "epoch": 0.5982947743874136, + "grad_norm": 0.0014049974270164967, + "learning_rate": 0.235826175269721, + "loss": 0.1179, + "num_input_tokens_seen": 15883520, + "step": 12245 + }, + { + "epoch": 0.5985390760511079, + "grad_norm": 0.0017737760208547115, + "learning_rate": 0.23577785889640612, + "loss": 0.0883, + "num_input_tokens_seen": 15889888, + "step": 12250 + }, + { + "epoch": 0.5987833777148023, + "grad_norm": 0.0006780466646887362, + "learning_rate": 0.23572952929506744, + "loss": 0.0905, + "num_input_tokens_seen": 15896448, + "step": 12255 + }, + { + "epoch": 0.5990276793784965, + "grad_norm": 0.0025699445977807045, + "learning_rate": 0.23568118647315803, + "loss": 0.1155, + "num_input_tokens_seen": 15902432, + "step": 12260 + }, + { + "epoch": 0.5992719810421909, + "grad_norm": 0.0015352084301412106, + "learning_rate": 0.23563283043813296, + "loss": 0.0958, + "num_input_tokens_seen": 15908992, + "step": 12265 + }, + { + "epoch": 0.5995162827058852, + "grad_norm": 0.0019590090960264206, + "learning_rate": 0.23558446119744922, + "loss": 0.1164, + "num_input_tokens_seen": 15915392, + "step": 12270 + }, + { + "epoch": 0.5997605843695796, + "grad_norm": 0.0015590678667649627, + "learning_rate": 0.23553607875856608, + "loss": 0.1178, + "num_input_tokens_seen": 15921728, + "step": 12275 + }, + { + "epoch": 0.6000048860332738, + "grad_norm": 0.0025057292077690363, + "learning_rate": 0.2354876831289447, + "loss": 0.1244, + "num_input_tokens_seen": 15928160, + "step": 12280 + }, + { + "epoch": 0.6002491876969682, + "grad_norm": 0.0016128862043842673, + "learning_rate": 0.23543927431604827, + "loss": 0.1193, + "num_input_tokens_seen": 15934464, + "step": 12285 + }, + { + "epoch": 0.6004934893606625, + "grad_norm": 0.001124630100093782, + "learning_rate": 0.23539085232734203, + "loss": 0.1213, + "num_input_tokens_seen": 15940800, + "step": 12290 + }, + { + "epoch": 0.6007377910243569, + "grad_norm": 0.0009270780719816685, + "learning_rate": 0.2353424171702933, + "loss": 0.139, + "num_input_tokens_seen": 15947296, + "step": 12295 + }, + { + "epoch": 0.6009820926880513, + "grad_norm": 0.002091828268021345, + "learning_rate": 0.23529396885237133, + "loss": 0.1152, + "num_input_tokens_seen": 15953664, + "step": 12300 + }, + { + "epoch": 0.6012263943517455, + "grad_norm": 0.0015733905602246523, + "learning_rate": 0.2352455073810475, + "loss": 0.1204, + "num_input_tokens_seen": 15960608, + "step": 12305 + }, + { + "epoch": 0.6014706960154399, + "grad_norm": 0.0008417971548624337, + "learning_rate": 0.23519703276379517, + "loss": 0.1181, + "num_input_tokens_seen": 15966912, + "step": 12310 + }, + { + "epoch": 0.6017149976791342, + "grad_norm": 0.0010291243670508265, + "learning_rate": 0.2351485450080897, + "loss": 0.1197, + "num_input_tokens_seen": 15973120, + "step": 12315 + }, + { + "epoch": 0.6019592993428285, + "grad_norm": 0.0012895339168608189, + "learning_rate": 0.2351000441214086, + "loss": 0.1016, + "num_input_tokens_seen": 15979360, + "step": 12320 + }, + { + "epoch": 0.6022036010065228, + "grad_norm": 0.001383866649121046, + "learning_rate": 0.23505153011123125, + "loss": 0.0889, + "num_input_tokens_seen": 15986400, + "step": 12325 + }, + { + "epoch": 0.6024479026702172, + "grad_norm": 0.0018827106105163693, + "learning_rate": 0.23500300298503912, + "loss": 0.1031, + "num_input_tokens_seen": 15992992, + "step": 12330 + }, + { + "epoch": 0.6026922043339115, + "grad_norm": 0.0018338961526751518, + "learning_rate": 0.23495446275031576, + "loss": 0.1212, + "num_input_tokens_seen": 15999040, + "step": 12335 + }, + { + "epoch": 0.6029365059976058, + "grad_norm": 0.0021866706665605307, + "learning_rate": 0.2349059094145466, + "loss": 0.1029, + "num_input_tokens_seen": 16006048, + "step": 12340 + }, + { + "epoch": 0.6031808076613002, + "grad_norm": 0.0021065999753773212, + "learning_rate": 0.2348573429852192, + "loss": 0.0975, + "num_input_tokens_seen": 16012224, + "step": 12345 + }, + { + "epoch": 0.6034251093249945, + "grad_norm": 0.002246052958071232, + "learning_rate": 0.23480876346982313, + "loss": 0.1201, + "num_input_tokens_seen": 16018816, + "step": 12350 + }, + { + "epoch": 0.6036694109886889, + "grad_norm": 0.002775737317278981, + "learning_rate": 0.23476017087585, + "loss": 0.1295, + "num_input_tokens_seen": 16025248, + "step": 12355 + }, + { + "epoch": 0.6039137126523831, + "grad_norm": 0.0012704615946859121, + "learning_rate": 0.23471156521079334, + "loss": 0.1443, + "num_input_tokens_seen": 16031744, + "step": 12360 + }, + { + "epoch": 0.6041580143160775, + "grad_norm": 0.0012831159401685, + "learning_rate": 0.23466294648214875, + "loss": 0.1182, + "num_input_tokens_seen": 16038944, + "step": 12365 + }, + { + "epoch": 0.6044023159797718, + "grad_norm": 0.0016806893981993198, + "learning_rate": 0.2346143146974139, + "loss": 0.1323, + "num_input_tokens_seen": 16045216, + "step": 12370 + }, + { + "epoch": 0.6046466176434662, + "grad_norm": 0.0006810103659518063, + "learning_rate": 0.23456566986408836, + "loss": 0.0951, + "num_input_tokens_seen": 16051584, + "step": 12375 + }, + { + "epoch": 0.6048909193071604, + "grad_norm": 0.001217854442074895, + "learning_rate": 0.23451701198967384, + "loss": 0.1193, + "num_input_tokens_seen": 16057664, + "step": 12380 + }, + { + "epoch": 0.6051352209708548, + "grad_norm": 0.0008935760124586523, + "learning_rate": 0.23446834108167397, + "loss": 0.1106, + "num_input_tokens_seen": 16064032, + "step": 12385 + }, + { + "epoch": 0.6053795226345492, + "grad_norm": 0.0013791992096230388, + "learning_rate": 0.23441965714759438, + "loss": 0.0914, + "num_input_tokens_seen": 16070528, + "step": 12390 + }, + { + "epoch": 0.6056238242982435, + "grad_norm": 0.0012658762279897928, + "learning_rate": 0.23437096019494277, + "loss": 0.1075, + "num_input_tokens_seen": 16076960, + "step": 12395 + }, + { + "epoch": 0.6058681259619378, + "grad_norm": 0.0015569081297144294, + "learning_rate": 0.23432225023122885, + "loss": 0.1099, + "num_input_tokens_seen": 16083104, + "step": 12400 + }, + { + "epoch": 0.6058681259619378, + "eval_loss": 0.11391448229551315, + "eval_runtime": 402.4417, + "eval_samples_per_second": 90.411, + "eval_steps_per_second": 22.605, + "num_input_tokens_seen": 16083104, + "step": 12400 + }, + { + "epoch": 0.6061124276256321, + "grad_norm": 0.0012908090138807893, + "learning_rate": 0.23427352726396428, + "loss": 0.0935, + "num_input_tokens_seen": 16089376, + "step": 12405 + }, + { + "epoch": 0.6063567292893265, + "grad_norm": 0.0014192218659445643, + "learning_rate": 0.2342247913006628, + "loss": 0.0745, + "num_input_tokens_seen": 16095808, + "step": 12410 + }, + { + "epoch": 0.6066010309530208, + "grad_norm": 0.0009670749423094094, + "learning_rate": 0.23417604234883999, + "loss": 0.1104, + "num_input_tokens_seen": 16102112, + "step": 12415 + }, + { + "epoch": 0.6068453326167151, + "grad_norm": 0.0011176335392519832, + "learning_rate": 0.23412728041601363, + "loss": 0.1031, + "num_input_tokens_seen": 16108960, + "step": 12420 + }, + { + "epoch": 0.6070896342804094, + "grad_norm": 0.0019523907685652375, + "learning_rate": 0.23407850550970347, + "loss": 0.1506, + "num_input_tokens_seen": 16115488, + "step": 12425 + }, + { + "epoch": 0.6073339359441038, + "grad_norm": 0.001731379539705813, + "learning_rate": 0.23402971763743116, + "loss": 0.1215, + "num_input_tokens_seen": 16121824, + "step": 12430 + }, + { + "epoch": 0.607578237607798, + "grad_norm": 0.0011134300148114562, + "learning_rate": 0.23398091680672037, + "loss": 0.0994, + "num_input_tokens_seen": 16128512, + "step": 12435 + }, + { + "epoch": 0.6078225392714924, + "grad_norm": 0.0008287748205475509, + "learning_rate": 0.23393210302509687, + "loss": 0.1024, + "num_input_tokens_seen": 16135264, + "step": 12440 + }, + { + "epoch": 0.6080668409351868, + "grad_norm": 0.0015408076578751206, + "learning_rate": 0.23388327630008832, + "loss": 0.1065, + "num_input_tokens_seen": 16142176, + "step": 12445 + }, + { + "epoch": 0.6083111425988811, + "grad_norm": 0.0010726078180596232, + "learning_rate": 0.23383443663922443, + "loss": 0.1069, + "num_input_tokens_seen": 16148832, + "step": 12450 + }, + { + "epoch": 0.6085554442625755, + "grad_norm": 0.0011842964449897408, + "learning_rate": 0.23378558405003685, + "loss": 0.0963, + "num_input_tokens_seen": 16154752, + "step": 12455 + }, + { + "epoch": 0.6087997459262697, + "grad_norm": 0.0008360505453310907, + "learning_rate": 0.2337367185400593, + "loss": 0.1037, + "num_input_tokens_seen": 16161344, + "step": 12460 + }, + { + "epoch": 0.6090440475899641, + "grad_norm": 0.0016866285586729646, + "learning_rate": 0.23368784011682747, + "loss": 0.1002, + "num_input_tokens_seen": 16167424, + "step": 12465 + }, + { + "epoch": 0.6092883492536584, + "grad_norm": 0.0013768522767350078, + "learning_rate": 0.23363894878787902, + "loss": 0.0925, + "num_input_tokens_seen": 16174112, + "step": 12470 + }, + { + "epoch": 0.6095326509173528, + "grad_norm": 0.0016268228646367788, + "learning_rate": 0.23359004456075352, + "loss": 0.14, + "num_input_tokens_seen": 16180640, + "step": 12475 + }, + { + "epoch": 0.609776952581047, + "grad_norm": 0.0010325873736292124, + "learning_rate": 0.23354112744299277, + "loss": 0.1168, + "num_input_tokens_seen": 16187104, + "step": 12480 + }, + { + "epoch": 0.6100212542447414, + "grad_norm": 0.001988989068195224, + "learning_rate": 0.2334921974421403, + "loss": 0.1244, + "num_input_tokens_seen": 16193280, + "step": 12485 + }, + { + "epoch": 0.6102655559084358, + "grad_norm": 0.0020813369192183018, + "learning_rate": 0.23344325456574178, + "loss": 0.1497, + "num_input_tokens_seen": 16199840, + "step": 12490 + }, + { + "epoch": 0.61050985757213, + "grad_norm": 0.0021927908528596163, + "learning_rate": 0.23339429882134477, + "loss": 0.1267, + "num_input_tokens_seen": 16206080, + "step": 12495 + }, + { + "epoch": 0.6107541592358244, + "grad_norm": 0.0012842605356127024, + "learning_rate": 0.23334533021649884, + "loss": 0.1144, + "num_input_tokens_seen": 16213152, + "step": 12500 + }, + { + "epoch": 0.6109984608995187, + "grad_norm": 0.0010611488251015544, + "learning_rate": 0.23329634875875566, + "loss": 0.1011, + "num_input_tokens_seen": 16219296, + "step": 12505 + }, + { + "epoch": 0.6112427625632131, + "grad_norm": 0.0010946044931188226, + "learning_rate": 0.23324735445566874, + "loss": 0.1158, + "num_input_tokens_seen": 16225792, + "step": 12510 + }, + { + "epoch": 0.6114870642269074, + "grad_norm": 0.001132210367359221, + "learning_rate": 0.2331983473147936, + "loss": 0.0978, + "num_input_tokens_seen": 16232576, + "step": 12515 + }, + { + "epoch": 0.6117313658906017, + "grad_norm": 0.0012123016640543938, + "learning_rate": 0.23314932734368776, + "loss": 0.1071, + "num_input_tokens_seen": 16239328, + "step": 12520 + }, + { + "epoch": 0.611975667554296, + "grad_norm": 0.0009508246439509094, + "learning_rate": 0.2331002945499107, + "loss": 0.1127, + "num_input_tokens_seen": 16245728, + "step": 12525 + }, + { + "epoch": 0.6122199692179904, + "grad_norm": 0.0010745598701760173, + "learning_rate": 0.23305124894102397, + "loss": 0.1078, + "num_input_tokens_seen": 16252288, + "step": 12530 + }, + { + "epoch": 0.6124642708816848, + "grad_norm": 0.0018324770499020815, + "learning_rate": 0.23300219052459092, + "loss": 0.1202, + "num_input_tokens_seen": 16258528, + "step": 12535 + }, + { + "epoch": 0.612708572545379, + "grad_norm": 0.0014496067306026816, + "learning_rate": 0.23295311930817708, + "loss": 0.1308, + "num_input_tokens_seen": 16264992, + "step": 12540 + }, + { + "epoch": 0.6129528742090734, + "grad_norm": 0.0012151874834671617, + "learning_rate": 0.23290403529934972, + "loss": 0.1027, + "num_input_tokens_seen": 16271296, + "step": 12545 + }, + { + "epoch": 0.6131971758727677, + "grad_norm": 0.0020848989952355623, + "learning_rate": 0.23285493850567832, + "loss": 0.1214, + "num_input_tokens_seen": 16277280, + "step": 12550 + }, + { + "epoch": 0.613441477536462, + "grad_norm": 0.0009628408588469028, + "learning_rate": 0.23280582893473414, + "loss": 0.0985, + "num_input_tokens_seen": 16283840, + "step": 12555 + }, + { + "epoch": 0.6136857792001563, + "grad_norm": 0.001002756180241704, + "learning_rate": 0.2327567065940906, + "loss": 0.0869, + "num_input_tokens_seen": 16290528, + "step": 12560 + }, + { + "epoch": 0.6139300808638507, + "grad_norm": 0.002822086913511157, + "learning_rate": 0.23270757149132285, + "loss": 0.131, + "num_input_tokens_seen": 16296928, + "step": 12565 + }, + { + "epoch": 0.614174382527545, + "grad_norm": 0.00163533806335181, + "learning_rate": 0.23265842363400827, + "loss": 0.1193, + "num_input_tokens_seen": 16303904, + "step": 12570 + }, + { + "epoch": 0.6144186841912394, + "grad_norm": 0.0011325915111228824, + "learning_rate": 0.23260926302972595, + "loss": 0.1084, + "num_input_tokens_seen": 16310496, + "step": 12575 + }, + { + "epoch": 0.6146629858549336, + "grad_norm": 0.0015836353413760662, + "learning_rate": 0.2325600896860572, + "loss": 0.1167, + "num_input_tokens_seen": 16316960, + "step": 12580 + }, + { + "epoch": 0.614907287518628, + "grad_norm": 0.0016632906626909971, + "learning_rate": 0.23251090361058505, + "loss": 0.1126, + "num_input_tokens_seen": 16323456, + "step": 12585 + }, + { + "epoch": 0.6151515891823224, + "grad_norm": 0.0010867482051253319, + "learning_rate": 0.23246170481089476, + "loss": 0.1201, + "num_input_tokens_seen": 16329952, + "step": 12590 + }, + { + "epoch": 0.6153958908460166, + "grad_norm": 0.0017031070310622454, + "learning_rate": 0.23241249329457317, + "loss": 0.0885, + "num_input_tokens_seen": 16336384, + "step": 12595 + }, + { + "epoch": 0.615640192509711, + "grad_norm": 0.0012889440404251218, + "learning_rate": 0.23236326906920957, + "loss": 0.1128, + "num_input_tokens_seen": 16342784, + "step": 12600 + }, + { + "epoch": 0.615640192509711, + "eval_loss": 0.10645465552806854, + "eval_runtime": 402.4213, + "eval_samples_per_second": 90.415, + "eval_steps_per_second": 22.606, + "num_input_tokens_seen": 16342784, + "step": 12600 + }, + { + "epoch": 0.6158844941734053, + "grad_norm": 0.0014068447053432465, + "learning_rate": 0.2323140321423948, + "loss": 0.0998, + "num_input_tokens_seen": 16349088, + "step": 12605 + }, + { + "epoch": 0.6161287958370997, + "grad_norm": 0.0020024715922772884, + "learning_rate": 0.23226478252172184, + "loss": 0.1095, + "num_input_tokens_seen": 16355456, + "step": 12610 + }, + { + "epoch": 0.6163730975007939, + "grad_norm": 0.0020538046956062317, + "learning_rate": 0.23221552021478561, + "loss": 0.0992, + "num_input_tokens_seen": 16361984, + "step": 12615 + }, + { + "epoch": 0.6166173991644883, + "grad_norm": 0.0010944042587652802, + "learning_rate": 0.232166245229183, + "loss": 0.0919, + "num_input_tokens_seen": 16368128, + "step": 12620 + }, + { + "epoch": 0.6168617008281826, + "grad_norm": 0.001425561378709972, + "learning_rate": 0.2321169575725128, + "loss": 0.087, + "num_input_tokens_seen": 16374720, + "step": 12625 + }, + { + "epoch": 0.617106002491877, + "grad_norm": 0.001858038012869656, + "learning_rate": 0.23206765725237577, + "loss": 0.1228, + "num_input_tokens_seen": 16380896, + "step": 12630 + }, + { + "epoch": 0.6173503041555713, + "grad_norm": 0.0015200490597635508, + "learning_rate": 0.2320183442763747, + "loss": 0.1174, + "num_input_tokens_seen": 16387296, + "step": 12635 + }, + { + "epoch": 0.6175946058192656, + "grad_norm": 0.0010653476929292083, + "learning_rate": 0.23196901865211422, + "loss": 0.0959, + "num_input_tokens_seen": 16394400, + "step": 12640 + }, + { + "epoch": 0.61783890748296, + "grad_norm": 0.001187328714877367, + "learning_rate": 0.231919680387201, + "loss": 0.1134, + "num_input_tokens_seen": 16400896, + "step": 12645 + }, + { + "epoch": 0.6180832091466543, + "grad_norm": 0.0014724974753335118, + "learning_rate": 0.23187032948924358, + "loss": 0.0945, + "num_input_tokens_seen": 16407616, + "step": 12650 + }, + { + "epoch": 0.6183275108103486, + "grad_norm": 0.0010999146616086364, + "learning_rate": 0.23182096596585247, + "loss": 0.1248, + "num_input_tokens_seen": 16414272, + "step": 12655 + }, + { + "epoch": 0.6185718124740429, + "grad_norm": 0.001234204857610166, + "learning_rate": 0.23177158982464025, + "loss": 0.0672, + "num_input_tokens_seen": 16421088, + "step": 12660 + }, + { + "epoch": 0.6188161141377373, + "grad_norm": 0.0014916668878868222, + "learning_rate": 0.23172220107322122, + "loss": 0.1042, + "num_input_tokens_seen": 16427392, + "step": 12665 + }, + { + "epoch": 0.6190604158014316, + "grad_norm": 0.000853530946187675, + "learning_rate": 0.23167279971921184, + "loss": 0.1144, + "num_input_tokens_seen": 16433440, + "step": 12670 + }, + { + "epoch": 0.6193047174651259, + "grad_norm": 0.0013385777128860354, + "learning_rate": 0.23162338577023034, + "loss": 0.1095, + "num_input_tokens_seen": 16440032, + "step": 12675 + }, + { + "epoch": 0.6195490191288203, + "grad_norm": 0.002221467671915889, + "learning_rate": 0.23157395923389704, + "loss": 0.116, + "num_input_tokens_seen": 16446688, + "step": 12680 + }, + { + "epoch": 0.6197933207925146, + "grad_norm": 0.0011005854466930032, + "learning_rate": 0.2315245201178341, + "loss": 0.103, + "num_input_tokens_seen": 16453184, + "step": 12685 + }, + { + "epoch": 0.620037622456209, + "grad_norm": 0.0019573981408029795, + "learning_rate": 0.23147506842966564, + "loss": 0.1201, + "num_input_tokens_seen": 16459936, + "step": 12690 + }, + { + "epoch": 0.6202819241199032, + "grad_norm": 0.0013534866739064455, + "learning_rate": 0.23142560417701774, + "loss": 0.084, + "num_input_tokens_seen": 16466240, + "step": 12695 + }, + { + "epoch": 0.6205262257835976, + "grad_norm": 0.0014923571143299341, + "learning_rate": 0.23137612736751845, + "loss": 0.1116, + "num_input_tokens_seen": 16472896, + "step": 12700 + }, + { + "epoch": 0.6207705274472919, + "grad_norm": 0.0011959545081481338, + "learning_rate": 0.23132663800879766, + "loss": 0.1165, + "num_input_tokens_seen": 16478848, + "step": 12705 + }, + { + "epoch": 0.6210148291109863, + "grad_norm": 0.0008801005315035582, + "learning_rate": 0.2312771361084873, + "loss": 0.1064, + "num_input_tokens_seen": 16485312, + "step": 12710 + }, + { + "epoch": 0.6212591307746805, + "grad_norm": 0.0013885198859497905, + "learning_rate": 0.23122762167422112, + "loss": 0.1197, + "num_input_tokens_seen": 16491552, + "step": 12715 + }, + { + "epoch": 0.6215034324383749, + "grad_norm": 0.0008954186923801899, + "learning_rate": 0.23117809471363493, + "loss": 0.0889, + "num_input_tokens_seen": 16498144, + "step": 12720 + }, + { + "epoch": 0.6217477341020692, + "grad_norm": 0.001668691635131836, + "learning_rate": 0.23112855523436637, + "loss": 0.0951, + "num_input_tokens_seen": 16504096, + "step": 12725 + }, + { + "epoch": 0.6219920357657636, + "grad_norm": 0.0012167154345661402, + "learning_rate": 0.23107900324405511, + "loss": 0.1083, + "num_input_tokens_seen": 16510656, + "step": 12730 + }, + { + "epoch": 0.6222363374294579, + "grad_norm": 0.0011426006676629186, + "learning_rate": 0.2310294387503426, + "loss": 0.118, + "num_input_tokens_seen": 16517120, + "step": 12735 + }, + { + "epoch": 0.6224806390931522, + "grad_norm": 0.0007488809642381966, + "learning_rate": 0.23097986176087237, + "loss": 0.0957, + "num_input_tokens_seen": 16523616, + "step": 12740 + }, + { + "epoch": 0.6227249407568466, + "grad_norm": 0.0009615637827664614, + "learning_rate": 0.23093027228328986, + "loss": 0.106, + "num_input_tokens_seen": 16529696, + "step": 12745 + }, + { + "epoch": 0.6229692424205409, + "grad_norm": 0.0007325618644244969, + "learning_rate": 0.23088067032524226, + "loss": 0.0726, + "num_input_tokens_seen": 16536064, + "step": 12750 + }, + { + "epoch": 0.6232135440842352, + "grad_norm": 0.002515033818781376, + "learning_rate": 0.23083105589437888, + "loss": 0.1184, + "num_input_tokens_seen": 16542848, + "step": 12755 + }, + { + "epoch": 0.6234578457479295, + "grad_norm": 0.0009235243778675795, + "learning_rate": 0.23078142899835094, + "loss": 0.1208, + "num_input_tokens_seen": 16549440, + "step": 12760 + }, + { + "epoch": 0.6237021474116239, + "grad_norm": 0.001003371668048203, + "learning_rate": 0.23073178964481147, + "loss": 0.1262, + "num_input_tokens_seen": 16556160, + "step": 12765 + }, + { + "epoch": 0.6239464490753182, + "grad_norm": 0.001218398567289114, + "learning_rate": 0.2306821378414155, + "loss": 0.1136, + "num_input_tokens_seen": 16562976, + "step": 12770 + }, + { + "epoch": 0.6241907507390125, + "grad_norm": 0.0015451700892299414, + "learning_rate": 0.2306324735958199, + "loss": 0.118, + "num_input_tokens_seen": 16569632, + "step": 12775 + }, + { + "epoch": 0.6244350524027069, + "grad_norm": 0.0012639230117201805, + "learning_rate": 0.23058279691568362, + "loss": 0.0891, + "num_input_tokens_seen": 16575840, + "step": 12780 + }, + { + "epoch": 0.6246793540664012, + "grad_norm": 0.0014723290223628283, + "learning_rate": 0.23053310780866745, + "loss": 0.1183, + "num_input_tokens_seen": 16582240, + "step": 12785 + }, + { + "epoch": 0.6249236557300956, + "grad_norm": 0.0014985466841608286, + "learning_rate": 0.23048340628243397, + "loss": 0.1053, + "num_input_tokens_seen": 16588608, + "step": 12790 + }, + { + "epoch": 0.6251679573937898, + "grad_norm": 0.0012799467658624053, + "learning_rate": 0.23043369234464783, + "loss": 0.1032, + "num_input_tokens_seen": 16595168, + "step": 12795 + }, + { + "epoch": 0.6254122590574842, + "grad_norm": 0.0010792043758556247, + "learning_rate": 0.2303839660029755, + "loss": 0.0954, + "num_input_tokens_seen": 16601824, + "step": 12800 + }, + { + "epoch": 0.6254122590574842, + "eval_loss": 0.1107550710439682, + "eval_runtime": 402.4441, + "eval_samples_per_second": 90.41, + "eval_steps_per_second": 22.604, + "num_input_tokens_seen": 16601824, + "step": 12800 + }, + { + "epoch": 0.6256565607211785, + "grad_norm": 0.0010198826203122735, + "learning_rate": 0.23033422726508548, + "loss": 0.1049, + "num_input_tokens_seen": 16607968, + "step": 12805 + }, + { + "epoch": 0.6259008623848729, + "grad_norm": 0.0013283143052831292, + "learning_rate": 0.23028447613864808, + "loss": 0.108, + "num_input_tokens_seen": 16613984, + "step": 12810 + }, + { + "epoch": 0.6261451640485671, + "grad_norm": 0.0017451042076572776, + "learning_rate": 0.2302347126313355, + "loss": 0.1013, + "num_input_tokens_seen": 16620160, + "step": 12815 + }, + { + "epoch": 0.6263894657122615, + "grad_norm": 0.0017517040250822902, + "learning_rate": 0.23018493675082197, + "loss": 0.1054, + "num_input_tokens_seen": 16626912, + "step": 12820 + }, + { + "epoch": 0.6266337673759559, + "grad_norm": 0.0022268127650022507, + "learning_rate": 0.2301351485047835, + "loss": 0.0922, + "num_input_tokens_seen": 16633472, + "step": 12825 + }, + { + "epoch": 0.6268780690396502, + "grad_norm": 0.0015398989198729396, + "learning_rate": 0.23008534790089813, + "loss": 0.1116, + "num_input_tokens_seen": 16639616, + "step": 12830 + }, + { + "epoch": 0.6271223707033445, + "grad_norm": 0.0013734705280512571, + "learning_rate": 0.2300355349468457, + "loss": 0.1233, + "num_input_tokens_seen": 16645792, + "step": 12835 + }, + { + "epoch": 0.6273666723670388, + "grad_norm": 0.0013185490388423204, + "learning_rate": 0.22998570965030793, + "loss": 0.1158, + "num_input_tokens_seen": 16652256, + "step": 12840 + }, + { + "epoch": 0.6276109740307332, + "grad_norm": 0.001497848890721798, + "learning_rate": 0.22993587201896862, + "loss": 0.1261, + "num_input_tokens_seen": 16658336, + "step": 12845 + }, + { + "epoch": 0.6278552756944275, + "grad_norm": 0.0012546360958367586, + "learning_rate": 0.2298860220605133, + "loss": 0.0887, + "num_input_tokens_seen": 16664640, + "step": 12850 + }, + { + "epoch": 0.6280995773581218, + "grad_norm": 0.0010489443084225059, + "learning_rate": 0.22983615978262942, + "loss": 0.1233, + "num_input_tokens_seen": 16671424, + "step": 12855 + }, + { + "epoch": 0.6283438790218161, + "grad_norm": 0.0010586235439404845, + "learning_rate": 0.22978628519300648, + "loss": 0.1071, + "num_input_tokens_seen": 16677856, + "step": 12860 + }, + { + "epoch": 0.6285881806855105, + "grad_norm": 0.001768506015650928, + "learning_rate": 0.22973639829933568, + "loss": 0.1172, + "num_input_tokens_seen": 16684256, + "step": 12865 + }, + { + "epoch": 0.6288324823492047, + "grad_norm": 0.0014042898546904325, + "learning_rate": 0.22968649910931027, + "loss": 0.0859, + "num_input_tokens_seen": 16690688, + "step": 12870 + }, + { + "epoch": 0.6290767840128991, + "grad_norm": 0.001371703576296568, + "learning_rate": 0.22963658763062528, + "loss": 0.1078, + "num_input_tokens_seen": 16697312, + "step": 12875 + }, + { + "epoch": 0.6293210856765935, + "grad_norm": 0.0010991463204845786, + "learning_rate": 0.22958666387097765, + "loss": 0.0868, + "num_input_tokens_seen": 16703360, + "step": 12880 + }, + { + "epoch": 0.6295653873402878, + "grad_norm": 0.0008378818165510893, + "learning_rate": 0.22953672783806633, + "loss": 0.1019, + "num_input_tokens_seen": 16709824, + "step": 12885 + }, + { + "epoch": 0.6298096890039822, + "grad_norm": 0.0010934779420495033, + "learning_rate": 0.22948677953959207, + "loss": 0.1209, + "num_input_tokens_seen": 16716256, + "step": 12890 + }, + { + "epoch": 0.6300539906676764, + "grad_norm": 0.0011933633359149098, + "learning_rate": 0.2294368189832575, + "loss": 0.1178, + "num_input_tokens_seen": 16722720, + "step": 12895 + }, + { + "epoch": 0.6302982923313708, + "grad_norm": 0.0016720844432711601, + "learning_rate": 0.2293868461767672, + "loss": 0.1032, + "num_input_tokens_seen": 16729504, + "step": 12900 + }, + { + "epoch": 0.6305425939950651, + "grad_norm": 0.0012119815219193697, + "learning_rate": 0.22933686112782758, + "loss": 0.1199, + "num_input_tokens_seen": 16735808, + "step": 12905 + }, + { + "epoch": 0.6307868956587595, + "grad_norm": 0.0014895603526383638, + "learning_rate": 0.22928686384414698, + "loss": 0.1143, + "num_input_tokens_seen": 16742880, + "step": 12910 + }, + { + "epoch": 0.6310311973224537, + "grad_norm": 0.0008453723276033998, + "learning_rate": 0.22923685433343552, + "loss": 0.0794, + "num_input_tokens_seen": 16749152, + "step": 12915 + }, + { + "epoch": 0.6312754989861481, + "grad_norm": 0.0022616719361394644, + "learning_rate": 0.22918683260340542, + "loss": 0.1059, + "num_input_tokens_seen": 16755712, + "step": 12920 + }, + { + "epoch": 0.6315198006498425, + "grad_norm": 0.0012098680017516017, + "learning_rate": 0.2291367986617706, + "loss": 0.0965, + "num_input_tokens_seen": 16762176, + "step": 12925 + }, + { + "epoch": 0.6317641023135367, + "grad_norm": 0.0018164002103731036, + "learning_rate": 0.22908675251624697, + "loss": 0.1157, + "num_input_tokens_seen": 16768672, + "step": 12930 + }, + { + "epoch": 0.6320084039772311, + "grad_norm": 0.0011526258895173669, + "learning_rate": 0.22903669417455216, + "loss": 0.1088, + "num_input_tokens_seen": 16774976, + "step": 12935 + }, + { + "epoch": 0.6322527056409254, + "grad_norm": 0.0010156650096178055, + "learning_rate": 0.22898662364440592, + "loss": 0.1102, + "num_input_tokens_seen": 16781408, + "step": 12940 + }, + { + "epoch": 0.6324970073046198, + "grad_norm": 0.0016187349101528525, + "learning_rate": 0.2289365409335297, + "loss": 0.1088, + "num_input_tokens_seen": 16788064, + "step": 12945 + }, + { + "epoch": 0.632741308968314, + "grad_norm": 0.0010766369523480535, + "learning_rate": 0.2288864460496469, + "loss": 0.1102, + "num_input_tokens_seen": 16794336, + "step": 12950 + }, + { + "epoch": 0.6329856106320084, + "grad_norm": 0.0009428428602404892, + "learning_rate": 0.22883633900048272, + "loss": 0.0875, + "num_input_tokens_seen": 16800992, + "step": 12955 + }, + { + "epoch": 0.6332299122957027, + "grad_norm": 0.0015927883796393871, + "learning_rate": 0.2287862197937644, + "loss": 0.1195, + "num_input_tokens_seen": 16807104, + "step": 12960 + }, + { + "epoch": 0.6334742139593971, + "grad_norm": 0.0016807891661301255, + "learning_rate": 0.2287360884372209, + "loss": 0.1066, + "num_input_tokens_seen": 16813664, + "step": 12965 + }, + { + "epoch": 0.6337185156230913, + "grad_norm": 0.0009956368012353778, + "learning_rate": 0.22868594493858307, + "loss": 0.0888, + "num_input_tokens_seen": 16819936, + "step": 12970 + }, + { + "epoch": 0.6339628172867857, + "grad_norm": 0.00247167213819921, + "learning_rate": 0.2286357893055837, + "loss": 0.0998, + "num_input_tokens_seen": 16825792, + "step": 12975 + }, + { + "epoch": 0.6342071189504801, + "grad_norm": 0.0015453778905794024, + "learning_rate": 0.22858562154595746, + "loss": 0.1261, + "num_input_tokens_seen": 16832704, + "step": 12980 + }, + { + "epoch": 0.6344514206141744, + "grad_norm": 0.0015505588380619884, + "learning_rate": 0.22853544166744078, + "loss": 0.1152, + "num_input_tokens_seen": 16839456, + "step": 12985 + }, + { + "epoch": 0.6346957222778687, + "grad_norm": 0.0020357731264084578, + "learning_rate": 0.22848524967777206, + "loss": 0.1171, + "num_input_tokens_seen": 16846752, + "step": 12990 + }, + { + "epoch": 0.634940023941563, + "grad_norm": 0.0009000770514830947, + "learning_rate": 0.22843504558469152, + "loss": 0.1156, + "num_input_tokens_seen": 16853600, + "step": 12995 + }, + { + "epoch": 0.6351843256052574, + "grad_norm": 0.001040993258357048, + "learning_rate": 0.2283848293959413, + "loss": 0.1, + "num_input_tokens_seen": 16860320, + "step": 13000 + }, + { + "epoch": 0.6351843256052574, + "eval_loss": 0.10784220695495605, + "eval_runtime": 401.9778, + "eval_samples_per_second": 90.515, + "eval_steps_per_second": 22.631, + "num_input_tokens_seen": 16860320, + "step": 13000 + }, + { + "epoch": 0.6354286272689517, + "grad_norm": 0.0010324025060981512, + "learning_rate": 0.22833460111926532, + "loss": 0.079, + "num_input_tokens_seen": 16867168, + "step": 13005 + }, + { + "epoch": 0.635672928932646, + "grad_norm": 0.001005431986413896, + "learning_rate": 0.22828436076240946, + "loss": 0.1064, + "num_input_tokens_seen": 16874080, + "step": 13010 + }, + { + "epoch": 0.6359172305963403, + "grad_norm": 0.0017886186251416802, + "learning_rate": 0.22823410833312135, + "loss": 0.1151, + "num_input_tokens_seen": 16880544, + "step": 13015 + }, + { + "epoch": 0.6361615322600347, + "grad_norm": 0.0013408809900283813, + "learning_rate": 0.2281838438391506, + "loss": 0.1377, + "num_input_tokens_seen": 16886944, + "step": 13020 + }, + { + "epoch": 0.6364058339237291, + "grad_norm": 0.001020760159008205, + "learning_rate": 0.22813356728824863, + "loss": 0.1411, + "num_input_tokens_seen": 16893280, + "step": 13025 + }, + { + "epoch": 0.6366501355874233, + "grad_norm": 0.001271530520170927, + "learning_rate": 0.2280832786881687, + "loss": 0.1236, + "num_input_tokens_seen": 16899456, + "step": 13030 + }, + { + "epoch": 0.6368944372511177, + "grad_norm": 0.001046549528837204, + "learning_rate": 0.22803297804666592, + "loss": 0.1013, + "num_input_tokens_seen": 16905856, + "step": 13035 + }, + { + "epoch": 0.637138738914812, + "grad_norm": 0.0007066029938869178, + "learning_rate": 0.22798266537149728, + "loss": 0.1007, + "num_input_tokens_seen": 16912064, + "step": 13040 + }, + { + "epoch": 0.6373830405785064, + "grad_norm": 0.001080637681297958, + "learning_rate": 0.22793234067042167, + "loss": 0.0891, + "num_input_tokens_seen": 16919008, + "step": 13045 + }, + { + "epoch": 0.6376273422422006, + "grad_norm": 0.0013834454584866762, + "learning_rate": 0.22788200395119979, + "loss": 0.1111, + "num_input_tokens_seen": 16925152, + "step": 13050 + }, + { + "epoch": 0.637871643905895, + "grad_norm": 0.0007426341762766242, + "learning_rate": 0.2278316552215942, + "loss": 0.0967, + "num_input_tokens_seen": 16931520, + "step": 13055 + }, + { + "epoch": 0.6381159455695893, + "grad_norm": 0.0010268660262227058, + "learning_rate": 0.22778129448936918, + "loss": 0.1226, + "num_input_tokens_seen": 16938208, + "step": 13060 + }, + { + "epoch": 0.6383602472332837, + "grad_norm": 0.0012327752774581313, + "learning_rate": 0.22773092176229118, + "loss": 0.113, + "num_input_tokens_seen": 16944704, + "step": 13065 + }, + { + "epoch": 0.638604548896978, + "grad_norm": 0.0006974324933253229, + "learning_rate": 0.22768053704812816, + "loss": 0.0829, + "num_input_tokens_seen": 16951360, + "step": 13070 + }, + { + "epoch": 0.6388488505606723, + "grad_norm": 0.0009440124267712235, + "learning_rate": 0.22763014035465018, + "loss": 0.092, + "num_input_tokens_seen": 16957856, + "step": 13075 + }, + { + "epoch": 0.6390931522243667, + "grad_norm": 0.0013328235363587737, + "learning_rate": 0.22757973168962892, + "loss": 0.1454, + "num_input_tokens_seen": 16963968, + "step": 13080 + }, + { + "epoch": 0.639337453888061, + "grad_norm": 0.0014734467258676887, + "learning_rate": 0.22752931106083818, + "loss": 0.1176, + "num_input_tokens_seen": 16970304, + "step": 13085 + }, + { + "epoch": 0.6395817555517553, + "grad_norm": 0.0013475227169692516, + "learning_rate": 0.22747887847605341, + "loss": 0.098, + "num_input_tokens_seen": 16976448, + "step": 13090 + }, + { + "epoch": 0.6398260572154496, + "grad_norm": 0.0009101863251999021, + "learning_rate": 0.22742843394305184, + "loss": 0.074, + "num_input_tokens_seen": 16983104, + "step": 13095 + }, + { + "epoch": 0.640070358879144, + "grad_norm": 0.0011720387265086174, + "learning_rate": 0.22737797746961272, + "loss": 0.0904, + "num_input_tokens_seen": 16989888, + "step": 13100 + }, + { + "epoch": 0.6403146605428383, + "grad_norm": 0.001385115785524249, + "learning_rate": 0.22732750906351712, + "loss": 0.0946, + "num_input_tokens_seen": 16996192, + "step": 13105 + }, + { + "epoch": 0.6405589622065326, + "grad_norm": 0.0011504818685352802, + "learning_rate": 0.22727702873254785, + "loss": 0.1336, + "num_input_tokens_seen": 17002400, + "step": 13110 + }, + { + "epoch": 0.6408032638702269, + "grad_norm": 0.0010812939144670963, + "learning_rate": 0.22722653648448968, + "loss": 0.1169, + "num_input_tokens_seen": 17008544, + "step": 13115 + }, + { + "epoch": 0.6410475655339213, + "grad_norm": 0.0011743069626390934, + "learning_rate": 0.22717603232712902, + "loss": 0.1074, + "num_input_tokens_seen": 17014880, + "step": 13120 + }, + { + "epoch": 0.6412918671976157, + "grad_norm": 0.0018980818567797542, + "learning_rate": 0.22712551626825436, + "loss": 0.1153, + "num_input_tokens_seen": 17021728, + "step": 13125 + }, + { + "epoch": 0.6415361688613099, + "grad_norm": 0.0009978050366044044, + "learning_rate": 0.2270749883156559, + "loss": 0.0907, + "num_input_tokens_seen": 17028000, + "step": 13130 + }, + { + "epoch": 0.6417804705250043, + "grad_norm": 0.001593540539033711, + "learning_rate": 0.22702444847712563, + "loss": 0.1107, + "num_input_tokens_seen": 17033952, + "step": 13135 + }, + { + "epoch": 0.6420247721886986, + "grad_norm": 0.001245293766260147, + "learning_rate": 0.22697389676045743, + "loss": 0.0974, + "num_input_tokens_seen": 17040448, + "step": 13140 + }, + { + "epoch": 0.642269073852393, + "grad_norm": 0.0017607141053304076, + "learning_rate": 0.22692333317344704, + "loss": 0.1304, + "num_input_tokens_seen": 17046464, + "step": 13145 + }, + { + "epoch": 0.6425133755160872, + "grad_norm": 0.001845150371082127, + "learning_rate": 0.22687275772389198, + "loss": 0.0977, + "num_input_tokens_seen": 17053248, + "step": 13150 + }, + { + "epoch": 0.6427576771797816, + "grad_norm": 0.0012117347214370966, + "learning_rate": 0.22682217041959168, + "loss": 0.113, + "num_input_tokens_seen": 17060352, + "step": 13155 + }, + { + "epoch": 0.6430019788434759, + "grad_norm": 0.0015206674579530954, + "learning_rate": 0.2267715712683473, + "loss": 0.1202, + "num_input_tokens_seen": 17066592, + "step": 13160 + }, + { + "epoch": 0.6432462805071703, + "grad_norm": 0.001275760238058865, + "learning_rate": 0.22672096027796182, + "loss": 0.1319, + "num_input_tokens_seen": 17073408, + "step": 13165 + }, + { + "epoch": 0.6434905821708646, + "grad_norm": 0.0010885512456297874, + "learning_rate": 0.22667033745624016, + "loss": 0.1208, + "num_input_tokens_seen": 17080064, + "step": 13170 + }, + { + "epoch": 0.6437348838345589, + "grad_norm": 0.0008459063828922808, + "learning_rate": 0.22661970281098895, + "loss": 0.0968, + "num_input_tokens_seen": 17086144, + "step": 13175 + }, + { + "epoch": 0.6439791854982533, + "grad_norm": 0.001400327542796731, + "learning_rate": 0.22656905635001667, + "loss": 0.1066, + "num_input_tokens_seen": 17092672, + "step": 13180 + }, + { + "epoch": 0.6442234871619476, + "grad_norm": 0.0013447654200717807, + "learning_rate": 0.2265183980811337, + "loss": 0.1011, + "num_input_tokens_seen": 17099008, + "step": 13185 + }, + { + "epoch": 0.6444677888256419, + "grad_norm": 0.0018909912323579192, + "learning_rate": 0.22646772801215218, + "loss": 0.1153, + "num_input_tokens_seen": 17105632, + "step": 13190 + }, + { + "epoch": 0.6447120904893362, + "grad_norm": 0.0009891639929264784, + "learning_rate": 0.22641704615088598, + "loss": 0.0985, + "num_input_tokens_seen": 17112192, + "step": 13195 + }, + { + "epoch": 0.6449563921530306, + "grad_norm": 0.0011885821586474776, + "learning_rate": 0.22636635250515103, + "loss": 0.1017, + "num_input_tokens_seen": 17118528, + "step": 13200 + }, + { + "epoch": 0.6449563921530306, + "eval_loss": 0.10704771429300308, + "eval_runtime": 402.1232, + "eval_samples_per_second": 90.482, + "eval_steps_per_second": 22.622, + "num_input_tokens_seen": 17118528, + "step": 13200 + }, + { + "epoch": 0.6452006938167248, + "grad_norm": 0.001454219571314752, + "learning_rate": 0.2263156470827648, + "loss": 0.1107, + "num_input_tokens_seen": 17124896, + "step": 13205 + }, + { + "epoch": 0.6454449954804192, + "grad_norm": 0.0017512041376903653, + "learning_rate": 0.22626492989154678, + "loss": 0.0763, + "num_input_tokens_seen": 17131520, + "step": 13210 + }, + { + "epoch": 0.6456892971441136, + "grad_norm": 0.0009341626428067684, + "learning_rate": 0.22621420093931813, + "loss": 0.0792, + "num_input_tokens_seen": 17137920, + "step": 13215 + }, + { + "epoch": 0.6459335988078079, + "grad_norm": 0.00201458390802145, + "learning_rate": 0.22616346023390194, + "loss": 0.12, + "num_input_tokens_seen": 17144288, + "step": 13220 + }, + { + "epoch": 0.6461779004715023, + "grad_norm": 0.0020580969285219908, + "learning_rate": 0.22611270778312306, + "loss": 0.1116, + "num_input_tokens_seen": 17150624, + "step": 13225 + }, + { + "epoch": 0.6464222021351965, + "grad_norm": 0.0010750090004876256, + "learning_rate": 0.2260619435948081, + "loss": 0.1031, + "num_input_tokens_seen": 17157344, + "step": 13230 + }, + { + "epoch": 0.6466665037988909, + "grad_norm": 0.0014998042024672031, + "learning_rate": 0.22601116767678567, + "loss": 0.0765, + "num_input_tokens_seen": 17164000, + "step": 13235 + }, + { + "epoch": 0.6469108054625852, + "grad_norm": 0.001619100570678711, + "learning_rate": 0.2259603800368859, + "loss": 0.1192, + "num_input_tokens_seen": 17170304, + "step": 13240 + }, + { + "epoch": 0.6471551071262795, + "grad_norm": 0.0015310043236240745, + "learning_rate": 0.22590958068294098, + "loss": 0.0914, + "num_input_tokens_seen": 17177056, + "step": 13245 + }, + { + "epoch": 0.6473994087899738, + "grad_norm": 0.0011903585400432348, + "learning_rate": 0.22585876962278478, + "loss": 0.0902, + "num_input_tokens_seen": 17183680, + "step": 13250 + }, + { + "epoch": 0.6476437104536682, + "grad_norm": 0.0011429025325924158, + "learning_rate": 0.22580794686425298, + "loss": 0.1139, + "num_input_tokens_seen": 17190112, + "step": 13255 + }, + { + "epoch": 0.6478880121173625, + "grad_norm": 0.0009385294979438186, + "learning_rate": 0.22575711241518312, + "loss": 0.0979, + "num_input_tokens_seen": 17196576, + "step": 13260 + }, + { + "epoch": 0.6481323137810568, + "grad_norm": 0.0016239730175584555, + "learning_rate": 0.22570626628341453, + "loss": 0.1097, + "num_input_tokens_seen": 17202528, + "step": 13265 + }, + { + "epoch": 0.6483766154447512, + "grad_norm": 0.0011000509839504957, + "learning_rate": 0.22565540847678828, + "loss": 0.1077, + "num_input_tokens_seen": 17209024, + "step": 13270 + }, + { + "epoch": 0.6486209171084455, + "grad_norm": 0.0012880222639068961, + "learning_rate": 0.2256045390031473, + "loss": 0.0979, + "num_input_tokens_seen": 17215424, + "step": 13275 + }, + { + "epoch": 0.6488652187721399, + "grad_norm": 0.00240444578230381, + "learning_rate": 0.22555365787033627, + "loss": 0.1494, + "num_input_tokens_seen": 17222176, + "step": 13280 + }, + { + "epoch": 0.6491095204358341, + "grad_norm": 0.0008635861449874938, + "learning_rate": 0.22550276508620173, + "loss": 0.0977, + "num_input_tokens_seen": 17229408, + "step": 13285 + }, + { + "epoch": 0.6493538220995285, + "grad_norm": 0.0012134192511439323, + "learning_rate": 0.22545186065859202, + "loss": 0.1067, + "num_input_tokens_seen": 17235936, + "step": 13290 + }, + { + "epoch": 0.6495981237632228, + "grad_norm": 0.0012667076662182808, + "learning_rate": 0.2254009445953572, + "loss": 0.0922, + "num_input_tokens_seen": 17242848, + "step": 13295 + }, + { + "epoch": 0.6498424254269172, + "grad_norm": 0.0009659474017098546, + "learning_rate": 0.22535001690434917, + "loss": 0.0859, + "num_input_tokens_seen": 17249408, + "step": 13300 + }, + { + "epoch": 0.6500867270906114, + "grad_norm": 0.0013993821339681745, + "learning_rate": 0.22529907759342163, + "loss": 0.0955, + "num_input_tokens_seen": 17255936, + "step": 13305 + }, + { + "epoch": 0.6503310287543058, + "grad_norm": 0.0021850892808288336, + "learning_rate": 0.22524812667043007, + "loss": 0.1302, + "num_input_tokens_seen": 17262624, + "step": 13310 + }, + { + "epoch": 0.6505753304180002, + "grad_norm": 0.0007365363417193294, + "learning_rate": 0.22519716414323177, + "loss": 0.0821, + "num_input_tokens_seen": 17269152, + "step": 13315 + }, + { + "epoch": 0.6508196320816945, + "grad_norm": 0.0021708612330257893, + "learning_rate": 0.22514619001968567, + "loss": 0.101, + "num_input_tokens_seen": 17275808, + "step": 13320 + }, + { + "epoch": 0.6510639337453888, + "grad_norm": 0.001336328568868339, + "learning_rate": 0.2250952043076528, + "loss": 0.1024, + "num_input_tokens_seen": 17282368, + "step": 13325 + }, + { + "epoch": 0.6513082354090831, + "grad_norm": 0.0018604568904265761, + "learning_rate": 0.2250442070149957, + "loss": 0.1012, + "num_input_tokens_seen": 17288864, + "step": 13330 + }, + { + "epoch": 0.6515525370727775, + "grad_norm": 0.0014160072896629572, + "learning_rate": 0.22499319814957885, + "loss": 0.1, + "num_input_tokens_seen": 17294976, + "step": 13335 + }, + { + "epoch": 0.6517968387364718, + "grad_norm": 0.0023832025472074747, + "learning_rate": 0.2249421777192684, + "loss": 0.1318, + "num_input_tokens_seen": 17301632, + "step": 13340 + }, + { + "epoch": 0.6520411404001661, + "grad_norm": 0.0009793041972443461, + "learning_rate": 0.22489114573193236, + "loss": 0.1029, + "num_input_tokens_seen": 17308000, + "step": 13345 + }, + { + "epoch": 0.6522854420638604, + "grad_norm": 0.002829594537615776, + "learning_rate": 0.2248401021954405, + "loss": 0.1039, + "num_input_tokens_seen": 17314208, + "step": 13350 + }, + { + "epoch": 0.6525297437275548, + "grad_norm": 0.001533425529487431, + "learning_rate": 0.22478904711766443, + "loss": 0.0884, + "num_input_tokens_seen": 17321216, + "step": 13355 + }, + { + "epoch": 0.6527740453912492, + "grad_norm": 0.002734034787863493, + "learning_rate": 0.22473798050647734, + "loss": 0.1059, + "num_input_tokens_seen": 17327840, + "step": 13360 + }, + { + "epoch": 0.6530183470549434, + "grad_norm": 0.0018205768428742886, + "learning_rate": 0.22468690236975453, + "loss": 0.1187, + "num_input_tokens_seen": 17334144, + "step": 13365 + }, + { + "epoch": 0.6532626487186378, + "grad_norm": 0.0018086680211126804, + "learning_rate": 0.22463581271537272, + "loss": 0.1295, + "num_input_tokens_seen": 17340448, + "step": 13370 + }, + { + "epoch": 0.6535069503823321, + "grad_norm": 0.0010361612075939775, + "learning_rate": 0.22458471155121076, + "loss": 0.1382, + "num_input_tokens_seen": 17346848, + "step": 13375 + }, + { + "epoch": 0.6537512520460265, + "grad_norm": 0.0016260507982224226, + "learning_rate": 0.2245335988851489, + "loss": 0.1074, + "num_input_tokens_seen": 17353216, + "step": 13380 + }, + { + "epoch": 0.6539955537097207, + "grad_norm": 0.0015133155975490808, + "learning_rate": 0.2244824747250695, + "loss": 0.1139, + "num_input_tokens_seen": 17359840, + "step": 13385 + }, + { + "epoch": 0.6542398553734151, + "grad_norm": 0.0012557184090837836, + "learning_rate": 0.22443133907885646, + "loss": 0.1333, + "num_input_tokens_seen": 17365376, + "step": 13390 + }, + { + "epoch": 0.6544841570371094, + "grad_norm": 0.0015820766566321254, + "learning_rate": 0.22438019195439557, + "loss": 0.1168, + "num_input_tokens_seen": 17372096, + "step": 13395 + }, + { + "epoch": 0.6547284587008038, + "grad_norm": 0.0010412579867988825, + "learning_rate": 0.22432903335957435, + "loss": 0.1085, + "num_input_tokens_seen": 17378528, + "step": 13400 + }, + { + "epoch": 0.6547284587008038, + "eval_loss": 0.10964423418045044, + "eval_runtime": 402.2496, + "eval_samples_per_second": 90.454, + "eval_steps_per_second": 22.615, + "num_input_tokens_seen": 17378528, + "step": 13400 + }, + { + "epoch": 0.654972760364498, + "grad_norm": 0.001981972483918071, + "learning_rate": 0.22427786330228214, + "loss": 0.1121, + "num_input_tokens_seen": 17384832, + "step": 13405 + }, + { + "epoch": 0.6552170620281924, + "grad_norm": 0.0011683354387059808, + "learning_rate": 0.22422668179040997, + "loss": 0.1043, + "num_input_tokens_seen": 17391360, + "step": 13410 + }, + { + "epoch": 0.6554613636918868, + "grad_norm": 0.0012274541659280658, + "learning_rate": 0.2241754888318507, + "loss": 0.1135, + "num_input_tokens_seen": 17397632, + "step": 13415 + }, + { + "epoch": 0.6557056653555811, + "grad_norm": 0.002088496694341302, + "learning_rate": 0.22412428443449886, + "loss": 0.1498, + "num_input_tokens_seen": 17404160, + "step": 13420 + }, + { + "epoch": 0.6559499670192754, + "grad_norm": 0.0019513457082211971, + "learning_rate": 0.22407306860625087, + "loss": 0.12, + "num_input_tokens_seen": 17410656, + "step": 13425 + }, + { + "epoch": 0.6561942686829697, + "grad_norm": 0.000779199181124568, + "learning_rate": 0.22402184135500483, + "loss": 0.0974, + "num_input_tokens_seen": 17416832, + "step": 13430 + }, + { + "epoch": 0.6564385703466641, + "grad_norm": 0.0010628425516188145, + "learning_rate": 0.22397060268866067, + "loss": 0.0995, + "num_input_tokens_seen": 17423104, + "step": 13435 + }, + { + "epoch": 0.6566828720103584, + "grad_norm": 0.0011732388520613313, + "learning_rate": 0.22391935261511994, + "loss": 0.1164, + "num_input_tokens_seen": 17429664, + "step": 13440 + }, + { + "epoch": 0.6569271736740527, + "grad_norm": 0.0013365171616896987, + "learning_rate": 0.22386809114228615, + "loss": 0.0936, + "num_input_tokens_seen": 17436096, + "step": 13445 + }, + { + "epoch": 0.657171475337747, + "grad_norm": 0.0011834672186523676, + "learning_rate": 0.22381681827806446, + "loss": 0.1254, + "num_input_tokens_seen": 17442688, + "step": 13450 + }, + { + "epoch": 0.6574157770014414, + "grad_norm": 0.0015049444045871496, + "learning_rate": 0.22376553403036173, + "loss": 0.1356, + "num_input_tokens_seen": 17449024, + "step": 13455 + }, + { + "epoch": 0.6576600786651358, + "grad_norm": 0.001222295337356627, + "learning_rate": 0.22371423840708662, + "loss": 0.1107, + "num_input_tokens_seen": 17455360, + "step": 13460 + }, + { + "epoch": 0.65790438032883, + "grad_norm": 0.0013284505112096667, + "learning_rate": 0.22366293141614962, + "loss": 0.125, + "num_input_tokens_seen": 17461632, + "step": 13465 + }, + { + "epoch": 0.6581486819925244, + "grad_norm": 0.0015695394249632955, + "learning_rate": 0.22361161306546287, + "loss": 0.096, + "num_input_tokens_seen": 17468416, + "step": 13470 + }, + { + "epoch": 0.6583929836562187, + "grad_norm": 0.0009431105572730303, + "learning_rate": 0.22356028336294037, + "loss": 0.1014, + "num_input_tokens_seen": 17474720, + "step": 13475 + }, + { + "epoch": 0.6586372853199131, + "grad_norm": 0.00106521334964782, + "learning_rate": 0.2235089423164977, + "loss": 0.1169, + "num_input_tokens_seen": 17481280, + "step": 13480 + }, + { + "epoch": 0.6588815869836073, + "grad_norm": 0.0008214006666094065, + "learning_rate": 0.22345758993405243, + "loss": 0.1113, + "num_input_tokens_seen": 17488032, + "step": 13485 + }, + { + "epoch": 0.6591258886473017, + "grad_norm": 0.001219106954522431, + "learning_rate": 0.2234062262235236, + "loss": 0.1396, + "num_input_tokens_seen": 17494272, + "step": 13490 + }, + { + "epoch": 0.659370190310996, + "grad_norm": 0.0012115027057006955, + "learning_rate": 0.22335485119283222, + "loss": 0.1009, + "num_input_tokens_seen": 17500544, + "step": 13495 + }, + { + "epoch": 0.6596144919746904, + "grad_norm": 0.0010024080984294415, + "learning_rate": 0.22330346484990093, + "loss": 0.1047, + "num_input_tokens_seen": 17506816, + "step": 13500 + }, + { + "epoch": 0.6598587936383846, + "grad_norm": 0.0011497748782858253, + "learning_rate": 0.22325206720265425, + "loss": 0.1133, + "num_input_tokens_seen": 17513536, + "step": 13505 + }, + { + "epoch": 0.660103095302079, + "grad_norm": 0.001236611045897007, + "learning_rate": 0.2232006582590182, + "loss": 0.1269, + "num_input_tokens_seen": 17520512, + "step": 13510 + }, + { + "epoch": 0.6603473969657734, + "grad_norm": 0.0010196688817813993, + "learning_rate": 0.22314923802692077, + "loss": 0.1088, + "num_input_tokens_seen": 17526944, + "step": 13515 + }, + { + "epoch": 0.6605916986294676, + "grad_norm": 0.0020244307816028595, + "learning_rate": 0.22309780651429156, + "loss": 0.0934, + "num_input_tokens_seen": 17533920, + "step": 13520 + }, + { + "epoch": 0.660836000293162, + "grad_norm": 0.0021185949444770813, + "learning_rate": 0.22304636372906203, + "loss": 0.1053, + "num_input_tokens_seen": 17540544, + "step": 13525 + }, + { + "epoch": 0.6610803019568563, + "grad_norm": 0.0022446548100560904, + "learning_rate": 0.22299490967916522, + "loss": 0.1104, + "num_input_tokens_seen": 17546880, + "step": 13530 + }, + { + "epoch": 0.6613246036205507, + "grad_norm": 0.0014564808225259185, + "learning_rate": 0.22294344437253602, + "loss": 0.0785, + "num_input_tokens_seen": 17554080, + "step": 13535 + }, + { + "epoch": 0.661568905284245, + "grad_norm": 0.0010907883988693357, + "learning_rate": 0.22289196781711101, + "loss": 0.0994, + "num_input_tokens_seen": 17560000, + "step": 13540 + }, + { + "epoch": 0.6618132069479393, + "grad_norm": 0.00094712054124102, + "learning_rate": 0.2228404800208286, + "loss": 0.0874, + "num_input_tokens_seen": 17566976, + "step": 13545 + }, + { + "epoch": 0.6620575086116336, + "grad_norm": 0.002135154092684388, + "learning_rate": 0.22278898099162875, + "loss": 0.1114, + "num_input_tokens_seen": 17573024, + "step": 13550 + }, + { + "epoch": 0.662301810275328, + "grad_norm": 0.0012776299845427275, + "learning_rate": 0.22273747073745337, + "loss": 0.1045, + "num_input_tokens_seen": 17579840, + "step": 13555 + }, + { + "epoch": 0.6625461119390224, + "grad_norm": 0.0008091671043075621, + "learning_rate": 0.22268594926624588, + "loss": 0.1205, + "num_input_tokens_seen": 17586432, + "step": 13560 + }, + { + "epoch": 0.6627904136027166, + "grad_norm": 0.0012158126337453723, + "learning_rate": 0.22263441658595162, + "loss": 0.1104, + "num_input_tokens_seen": 17592640, + "step": 13565 + }, + { + "epoch": 0.663034715266411, + "grad_norm": 0.0010563336545601487, + "learning_rate": 0.2225828727045175, + "loss": 0.0979, + "num_input_tokens_seen": 17598976, + "step": 13570 + }, + { + "epoch": 0.6632790169301053, + "grad_norm": 0.0008653172408230603, + "learning_rate": 0.22253131762989228, + "loss": 0.0875, + "num_input_tokens_seen": 17605920, + "step": 13575 + }, + { + "epoch": 0.6635233185937996, + "grad_norm": 0.0027100425213575363, + "learning_rate": 0.2224797513700264, + "loss": 0.146, + "num_input_tokens_seen": 17612544, + "step": 13580 + }, + { + "epoch": 0.6637676202574939, + "grad_norm": 0.0007373004918918014, + "learning_rate": 0.22242817393287204, + "loss": 0.129, + "num_input_tokens_seen": 17618944, + "step": 13585 + }, + { + "epoch": 0.6640119219211883, + "grad_norm": 0.0008181840530596673, + "learning_rate": 0.22237658532638305, + "loss": 0.0824, + "num_input_tokens_seen": 17625696, + "step": 13590 + }, + { + "epoch": 0.6642562235848826, + "grad_norm": 0.0012951624812558293, + "learning_rate": 0.22232498555851513, + "loss": 0.1079, + "num_input_tokens_seen": 17632160, + "step": 13595 + }, + { + "epoch": 0.664500525248577, + "grad_norm": 0.0008909399621188641, + "learning_rate": 0.22227337463722546, + "loss": 0.0944, + "num_input_tokens_seen": 17638400, + "step": 13600 + }, + { + "epoch": 0.664500525248577, + "eval_loss": 0.1083836480975151, + "eval_runtime": 402.4333, + "eval_samples_per_second": 90.412, + "eval_steps_per_second": 22.605, + "num_input_tokens_seen": 17638400, + "step": 13600 + }, + { + "epoch": 0.6647448269122713, + "grad_norm": 0.0011121879797428846, + "learning_rate": 0.2222217525704732, + "loss": 0.0983, + "num_input_tokens_seen": 17644832, + "step": 13605 + }, + { + "epoch": 0.6649891285759656, + "grad_norm": 0.0013825857313349843, + "learning_rate": 0.22217011936621908, + "loss": 0.1302, + "num_input_tokens_seen": 17651424, + "step": 13610 + }, + { + "epoch": 0.66523343023966, + "grad_norm": 0.0013366761850193143, + "learning_rate": 0.22211847503242566, + "loss": 0.1123, + "num_input_tokens_seen": 17657728, + "step": 13615 + }, + { + "epoch": 0.6654777319033542, + "grad_norm": 0.0024751608725637197, + "learning_rate": 0.22206681957705704, + "loss": 0.108, + "num_input_tokens_seen": 17664064, + "step": 13620 + }, + { + "epoch": 0.6657220335670486, + "grad_norm": 0.0007183079142123461, + "learning_rate": 0.2220151530080792, + "loss": 0.0984, + "num_input_tokens_seen": 17671104, + "step": 13625 + }, + { + "epoch": 0.6659663352307429, + "grad_norm": 0.0007610308821313083, + "learning_rate": 0.2219634753334598, + "loss": 0.0858, + "num_input_tokens_seen": 17677696, + "step": 13630 + }, + { + "epoch": 0.6662106368944373, + "grad_norm": 0.001505788299255073, + "learning_rate": 0.22191178656116817, + "loss": 0.1168, + "num_input_tokens_seen": 17684000, + "step": 13635 + }, + { + "epoch": 0.6664549385581315, + "grad_norm": 0.0015417286194860935, + "learning_rate": 0.2218600866991753, + "loss": 0.1046, + "num_input_tokens_seen": 17690624, + "step": 13640 + }, + { + "epoch": 0.6666992402218259, + "grad_norm": 0.0010594673221930861, + "learning_rate": 0.221808375755454, + "loss": 0.0807, + "num_input_tokens_seen": 17696992, + "step": 13645 + }, + { + "epoch": 0.6669435418855202, + "grad_norm": 0.0022226907312870026, + "learning_rate": 0.22175665373797881, + "loss": 0.1067, + "num_input_tokens_seen": 17703296, + "step": 13650 + }, + { + "epoch": 0.6671878435492146, + "grad_norm": 0.00075796979945153, + "learning_rate": 0.22170492065472583, + "loss": 0.1013, + "num_input_tokens_seen": 17709888, + "step": 13655 + }, + { + "epoch": 0.667432145212909, + "grad_norm": 0.0016083925729617476, + "learning_rate": 0.221653176513673, + "loss": 0.1252, + "num_input_tokens_seen": 17716416, + "step": 13660 + }, + { + "epoch": 0.6676764468766032, + "grad_norm": 0.0012794187059625983, + "learning_rate": 0.2216014213227999, + "loss": 0.1096, + "num_input_tokens_seen": 17722752, + "step": 13665 + }, + { + "epoch": 0.6679207485402976, + "grad_norm": 0.002664062660187483, + "learning_rate": 0.22154965509008784, + "loss": 0.1332, + "num_input_tokens_seen": 17728992, + "step": 13670 + }, + { + "epoch": 0.6681650502039919, + "grad_norm": 0.0012936415150761604, + "learning_rate": 0.2214978778235198, + "loss": 0.0882, + "num_input_tokens_seen": 17735456, + "step": 13675 + }, + { + "epoch": 0.6684093518676862, + "grad_norm": 0.0011098046088591218, + "learning_rate": 0.2214460895310805, + "loss": 0.0903, + "num_input_tokens_seen": 17742176, + "step": 13680 + }, + { + "epoch": 0.6686536535313805, + "grad_norm": 0.0010337420972064137, + "learning_rate": 0.22139429022075635, + "loss": 0.1036, + "num_input_tokens_seen": 17748672, + "step": 13685 + }, + { + "epoch": 0.6688979551950749, + "grad_norm": 0.002051117131486535, + "learning_rate": 0.22134247990053546, + "loss": 0.1133, + "num_input_tokens_seen": 17754880, + "step": 13690 + }, + { + "epoch": 0.6691422568587692, + "grad_norm": 0.0010683774016797543, + "learning_rate": 0.2212906585784076, + "loss": 0.1043, + "num_input_tokens_seen": 17761344, + "step": 13695 + }, + { + "epoch": 0.6693865585224635, + "grad_norm": 0.001448330469429493, + "learning_rate": 0.22123882626236432, + "loss": 0.0866, + "num_input_tokens_seen": 17767616, + "step": 13700 + }, + { + "epoch": 0.6696308601861579, + "grad_norm": 0.0007619251264259219, + "learning_rate": 0.2211869829603988, + "loss": 0.0864, + "num_input_tokens_seen": 17774048, + "step": 13705 + }, + { + "epoch": 0.6698751618498522, + "grad_norm": 0.001402788795530796, + "learning_rate": 0.22113512868050592, + "loss": 0.0953, + "num_input_tokens_seen": 17780576, + "step": 13710 + }, + { + "epoch": 0.6701194635135466, + "grad_norm": 0.002695039613172412, + "learning_rate": 0.2210832634306822, + "loss": 0.1119, + "num_input_tokens_seen": 17787456, + "step": 13715 + }, + { + "epoch": 0.6703637651772408, + "grad_norm": 0.0012250670697540045, + "learning_rate": 0.22103138721892598, + "loss": 0.1049, + "num_input_tokens_seen": 17794272, + "step": 13720 + }, + { + "epoch": 0.6706080668409352, + "grad_norm": 0.0014838054776191711, + "learning_rate": 0.22097950005323724, + "loss": 0.0823, + "num_input_tokens_seen": 17801120, + "step": 13725 + }, + { + "epoch": 0.6708523685046295, + "grad_norm": 0.0023204919416457415, + "learning_rate": 0.22092760194161762, + "loss": 0.1206, + "num_input_tokens_seen": 17807680, + "step": 13730 + }, + { + "epoch": 0.6710966701683239, + "grad_norm": 0.0015605330700054765, + "learning_rate": 0.2208756928920704, + "loss": 0.0959, + "num_input_tokens_seen": 17814400, + "step": 13735 + }, + { + "epoch": 0.6713409718320181, + "grad_norm": 0.0012490213848650455, + "learning_rate": 0.22082377291260072, + "loss": 0.1083, + "num_input_tokens_seen": 17820960, + "step": 13740 + }, + { + "epoch": 0.6715852734957125, + "grad_norm": 0.0009998263558372855, + "learning_rate": 0.2207718420112152, + "loss": 0.102, + "num_input_tokens_seen": 17827200, + "step": 13745 + }, + { + "epoch": 0.6718295751594069, + "grad_norm": 0.0019195060012862086, + "learning_rate": 0.22071990019592228, + "loss": 0.1068, + "num_input_tokens_seen": 17833824, + "step": 13750 + }, + { + "epoch": 0.6720738768231012, + "grad_norm": 0.0009505474590696394, + "learning_rate": 0.22066794747473198, + "loss": 0.0865, + "num_input_tokens_seen": 17840640, + "step": 13755 + }, + { + "epoch": 0.6723181784867955, + "grad_norm": 0.0016481016064062715, + "learning_rate": 0.2206159838556562, + "loss": 0.0946, + "num_input_tokens_seen": 17847264, + "step": 13760 + }, + { + "epoch": 0.6725624801504898, + "grad_norm": 0.0011278500314801931, + "learning_rate": 0.2205640093467082, + "loss": 0.0863, + "num_input_tokens_seen": 17853888, + "step": 13765 + }, + { + "epoch": 0.6728067818141842, + "grad_norm": 0.0014681399334222078, + "learning_rate": 0.22051202395590322, + "loss": 0.0811, + "num_input_tokens_seen": 17860544, + "step": 13770 + }, + { + "epoch": 0.6730510834778785, + "grad_norm": 0.001853914582170546, + "learning_rate": 0.22046002769125808, + "loss": 0.082, + "num_input_tokens_seen": 17866720, + "step": 13775 + }, + { + "epoch": 0.6732953851415728, + "grad_norm": 0.0012671623844653368, + "learning_rate": 0.2204080205607912, + "loss": 0.0909, + "num_input_tokens_seen": 17873248, + "step": 13780 + }, + { + "epoch": 0.6735396868052671, + "grad_norm": 0.001893278444185853, + "learning_rate": 0.22035600257252272, + "loss": 0.1089, + "num_input_tokens_seen": 17879616, + "step": 13785 + }, + { + "epoch": 0.6737839884689615, + "grad_norm": 0.001233745482750237, + "learning_rate": 0.2203039737344745, + "loss": 0.1353, + "num_input_tokens_seen": 17885856, + "step": 13790 + }, + { + "epoch": 0.6740282901326557, + "grad_norm": 0.0011282083578407764, + "learning_rate": 0.22025193405467003, + "loss": 0.141, + "num_input_tokens_seen": 17892352, + "step": 13795 + }, + { + "epoch": 0.6742725917963501, + "grad_norm": 0.0010935304453596473, + "learning_rate": 0.2201998835411345, + "loss": 0.0976, + "num_input_tokens_seen": 17898336, + "step": 13800 + }, + { + "epoch": 0.6742725917963501, + "eval_loss": 0.10950179398059845, + "eval_runtime": 402.577, + "eval_samples_per_second": 90.38, + "eval_steps_per_second": 22.597, + "num_input_tokens_seen": 17898336, + "step": 13800 + }, + { + "epoch": 0.6745168934600445, + "grad_norm": 0.001246976898983121, + "learning_rate": 0.22014782220189474, + "loss": 0.1047, + "num_input_tokens_seen": 17905184, + "step": 13805 + }, + { + "epoch": 0.6747611951237388, + "grad_norm": 0.0011533783981576562, + "learning_rate": 0.2200957500449793, + "loss": 0.1129, + "num_input_tokens_seen": 17911488, + "step": 13810 + }, + { + "epoch": 0.6750054967874332, + "grad_norm": 0.0012805608566850424, + "learning_rate": 0.22004366707841827, + "loss": 0.1088, + "num_input_tokens_seen": 17918048, + "step": 13815 + }, + { + "epoch": 0.6752497984511274, + "grad_norm": 0.0017085091676563025, + "learning_rate": 0.21999157331024358, + "loss": 0.1074, + "num_input_tokens_seen": 17924736, + "step": 13820 + }, + { + "epoch": 0.6754941001148218, + "grad_norm": 0.0014364084927365184, + "learning_rate": 0.21993946874848871, + "loss": 0.0951, + "num_input_tokens_seen": 17931072, + "step": 13825 + }, + { + "epoch": 0.6757384017785161, + "grad_norm": 0.0009540438186377287, + "learning_rate": 0.2198873534011888, + "loss": 0.0819, + "num_input_tokens_seen": 17937280, + "step": 13830 + }, + { + "epoch": 0.6759827034422105, + "grad_norm": 0.001071352162398398, + "learning_rate": 0.2198352272763808, + "loss": 0.0875, + "num_input_tokens_seen": 17943776, + "step": 13835 + }, + { + "epoch": 0.6762270051059047, + "grad_norm": 0.0019146567210555077, + "learning_rate": 0.2197830903821031, + "loss": 0.1245, + "num_input_tokens_seen": 17950496, + "step": 13840 + }, + { + "epoch": 0.6764713067695991, + "grad_norm": 0.0010987073183059692, + "learning_rate": 0.21973094272639598, + "loss": 0.0919, + "num_input_tokens_seen": 17956704, + "step": 13845 + }, + { + "epoch": 0.6767156084332935, + "grad_norm": 0.0019455419387668371, + "learning_rate": 0.21967878431730117, + "loss": 0.0926, + "num_input_tokens_seen": 17963200, + "step": 13850 + }, + { + "epoch": 0.6769599100969877, + "grad_norm": 0.0015561673790216446, + "learning_rate": 0.21962661516286217, + "loss": 0.1047, + "num_input_tokens_seen": 17970272, + "step": 13855 + }, + { + "epoch": 0.6772042117606821, + "grad_norm": 0.0012245442485436797, + "learning_rate": 0.21957443527112414, + "loss": 0.0981, + "num_input_tokens_seen": 17977088, + "step": 13860 + }, + { + "epoch": 0.6774485134243764, + "grad_norm": 0.001161319320090115, + "learning_rate": 0.21952224465013384, + "loss": 0.1096, + "num_input_tokens_seen": 17983552, + "step": 13865 + }, + { + "epoch": 0.6776928150880708, + "grad_norm": 0.0019134279573336244, + "learning_rate": 0.21947004330793976, + "loss": 0.0878, + "num_input_tokens_seen": 17990272, + "step": 13870 + }, + { + "epoch": 0.677937116751765, + "grad_norm": 0.0030622149351984262, + "learning_rate": 0.21941783125259198, + "loss": 0.0848, + "num_input_tokens_seen": 17996832, + "step": 13875 + }, + { + "epoch": 0.6781814184154594, + "grad_norm": 0.001772688003256917, + "learning_rate": 0.21936560849214226, + "loss": 0.1553, + "num_input_tokens_seen": 18003360, + "step": 13880 + }, + { + "epoch": 0.6784257200791537, + "grad_norm": 0.0008142695296555758, + "learning_rate": 0.21931337503464404, + "loss": 0.1013, + "num_input_tokens_seen": 18009984, + "step": 13885 + }, + { + "epoch": 0.6786700217428481, + "grad_norm": 0.001217205310240388, + "learning_rate": 0.21926113088815233, + "loss": 0.0997, + "num_input_tokens_seen": 18016448, + "step": 13890 + }, + { + "epoch": 0.6789143234065425, + "grad_norm": 0.001349880825728178, + "learning_rate": 0.2192088760607238, + "loss": 0.1076, + "num_input_tokens_seen": 18023232, + "step": 13895 + }, + { + "epoch": 0.6791586250702367, + "grad_norm": 0.0010120023507624865, + "learning_rate": 0.2191566105604169, + "loss": 0.1065, + "num_input_tokens_seen": 18029408, + "step": 13900 + }, + { + "epoch": 0.6794029267339311, + "grad_norm": 0.001077753957360983, + "learning_rate": 0.21910433439529153, + "loss": 0.1079, + "num_input_tokens_seen": 18035808, + "step": 13905 + }, + { + "epoch": 0.6796472283976254, + "grad_norm": 0.001729518175125122, + "learning_rate": 0.2190520475734094, + "loss": 0.087, + "num_input_tokens_seen": 18042784, + "step": 13910 + }, + { + "epoch": 0.6798915300613197, + "grad_norm": 0.0016491685528308153, + "learning_rate": 0.2189997501028338, + "loss": 0.1127, + "num_input_tokens_seen": 18049184, + "step": 13915 + }, + { + "epoch": 0.680135831725014, + "grad_norm": 0.0016279882984235883, + "learning_rate": 0.2189474419916296, + "loss": 0.1297, + "num_input_tokens_seen": 18055360, + "step": 13920 + }, + { + "epoch": 0.6803801333887084, + "grad_norm": 0.001870027044788003, + "learning_rate": 0.21889512324786342, + "loss": 0.0957, + "num_input_tokens_seen": 18061856, + "step": 13925 + }, + { + "epoch": 0.6806244350524027, + "grad_norm": 0.0013982156524434686, + "learning_rate": 0.21884279387960345, + "loss": 0.122, + "num_input_tokens_seen": 18068544, + "step": 13930 + }, + { + "epoch": 0.680868736716097, + "grad_norm": 0.0014774786541238427, + "learning_rate": 0.2187904538949195, + "loss": 0.1412, + "num_input_tokens_seen": 18074944, + "step": 13935 + }, + { + "epoch": 0.6811130383797913, + "grad_norm": 0.0012672393349930644, + "learning_rate": 0.2187381033018831, + "loss": 0.1291, + "num_input_tokens_seen": 18081248, + "step": 13940 + }, + { + "epoch": 0.6813573400434857, + "grad_norm": 0.002069746144115925, + "learning_rate": 0.2186857421085673, + "loss": 0.1426, + "num_input_tokens_seen": 18087552, + "step": 13945 + }, + { + "epoch": 0.6816016417071801, + "grad_norm": 0.0019475676817819476, + "learning_rate": 0.21863337032304697, + "loss": 0.1402, + "num_input_tokens_seen": 18093632, + "step": 13950 + }, + { + "epoch": 0.6818459433708743, + "grad_norm": 0.0014879830414429307, + "learning_rate": 0.21858098795339845, + "loss": 0.1219, + "num_input_tokens_seen": 18099904, + "step": 13955 + }, + { + "epoch": 0.6820902450345687, + "grad_norm": 0.0012946341885253787, + "learning_rate": 0.21852859500769975, + "loss": 0.1049, + "num_input_tokens_seen": 18106400, + "step": 13960 + }, + { + "epoch": 0.682334546698263, + "grad_norm": 0.0010259003611281514, + "learning_rate": 0.21847619149403044, + "loss": 0.1016, + "num_input_tokens_seen": 18113440, + "step": 13965 + }, + { + "epoch": 0.6825788483619574, + "grad_norm": 0.0010437039891257882, + "learning_rate": 0.21842377742047195, + "loss": 0.105, + "num_input_tokens_seen": 18120064, + "step": 13970 + }, + { + "epoch": 0.6828231500256516, + "grad_norm": 0.001040846575051546, + "learning_rate": 0.21837135279510705, + "loss": 0.1079, + "num_input_tokens_seen": 18126464, + "step": 13975 + }, + { + "epoch": 0.683067451689346, + "grad_norm": 0.0010404333006590605, + "learning_rate": 0.21831891762602038, + "loss": 0.1161, + "num_input_tokens_seen": 18133184, + "step": 13980 + }, + { + "epoch": 0.6833117533530403, + "grad_norm": 0.001912300125695765, + "learning_rate": 0.21826647192129806, + "loss": 0.1284, + "num_input_tokens_seen": 18139264, + "step": 13985 + }, + { + "epoch": 0.6835560550167347, + "grad_norm": 0.001052029780112207, + "learning_rate": 0.21821401568902787, + "loss": 0.103, + "num_input_tokens_seen": 18145728, + "step": 13990 + }, + { + "epoch": 0.683800356680429, + "grad_norm": 0.0016341995215043426, + "learning_rate": 0.21816154893729925, + "loss": 0.1215, + "num_input_tokens_seen": 18152032, + "step": 13995 + }, + { + "epoch": 0.6840446583441233, + "grad_norm": 0.0009057443821802735, + "learning_rate": 0.2181090716742032, + "loss": 0.1022, + "num_input_tokens_seen": 18158528, + "step": 14000 + }, + { + "epoch": 0.6840446583441233, + "eval_loss": 0.11265753209590912, + "eval_runtime": 402.2397, + "eval_samples_per_second": 90.456, + "eval_steps_per_second": 22.616, + "num_input_tokens_seen": 18158528, + "step": 14000 + }, + { + "epoch": 0.6842889600078177, + "grad_norm": 0.0015836501261219382, + "learning_rate": 0.21805658390783236, + "loss": 0.125, + "num_input_tokens_seen": 18164768, + "step": 14005 + }, + { + "epoch": 0.684533261671512, + "grad_norm": 0.0013598310761153698, + "learning_rate": 0.21800408564628107, + "loss": 0.1142, + "num_input_tokens_seen": 18171648, + "step": 14010 + }, + { + "epoch": 0.6847775633352063, + "grad_norm": 0.0007829886744730175, + "learning_rate": 0.21795157689764516, + "loss": 0.1046, + "num_input_tokens_seen": 18178016, + "step": 14015 + }, + { + "epoch": 0.6850218649989006, + "grad_norm": 0.0011518746614456177, + "learning_rate": 0.21789905767002216, + "loss": 0.1056, + "num_input_tokens_seen": 18184768, + "step": 14020 + }, + { + "epoch": 0.685266166662595, + "grad_norm": 0.001317039830610156, + "learning_rate": 0.2178465279715112, + "loss": 0.0785, + "num_input_tokens_seen": 18191328, + "step": 14025 + }, + { + "epoch": 0.6855104683262893, + "grad_norm": 0.0018145197536796331, + "learning_rate": 0.21779398781021303, + "loss": 0.1334, + "num_input_tokens_seen": 18197760, + "step": 14030 + }, + { + "epoch": 0.6857547699899836, + "grad_norm": 0.0009556583245284855, + "learning_rate": 0.21774143719422998, + "loss": 0.1079, + "num_input_tokens_seen": 18204128, + "step": 14035 + }, + { + "epoch": 0.685999071653678, + "grad_norm": 0.0011485617142170668, + "learning_rate": 0.21768887613166601, + "loss": 0.1112, + "num_input_tokens_seen": 18210336, + "step": 14040 + }, + { + "epoch": 0.6862433733173723, + "grad_norm": 0.001054224674589932, + "learning_rate": 0.2176363046306267, + "loss": 0.0907, + "num_input_tokens_seen": 18217376, + "step": 14045 + }, + { + "epoch": 0.6864876749810667, + "grad_norm": 0.002937370678409934, + "learning_rate": 0.21758372269921925, + "loss": 0.1061, + "num_input_tokens_seen": 18223616, + "step": 14050 + }, + { + "epoch": 0.6867319766447609, + "grad_norm": 0.002940363949164748, + "learning_rate": 0.21753113034555244, + "loss": 0.1022, + "num_input_tokens_seen": 18230368, + "step": 14055 + }, + { + "epoch": 0.6869762783084553, + "grad_norm": 0.002577039413154125, + "learning_rate": 0.2174785275777367, + "loss": 0.1096, + "num_input_tokens_seen": 18236544, + "step": 14060 + }, + { + "epoch": 0.6872205799721496, + "grad_norm": 0.0015164174837991595, + "learning_rate": 0.21742591440388404, + "loss": 0.098, + "num_input_tokens_seen": 18243264, + "step": 14065 + }, + { + "epoch": 0.687464881635844, + "grad_norm": 0.0007711936486884952, + "learning_rate": 0.21737329083210802, + "loss": 0.0869, + "num_input_tokens_seen": 18249632, + "step": 14070 + }, + { + "epoch": 0.6877091832995382, + "grad_norm": 0.0008166284533217549, + "learning_rate": 0.2173206568705239, + "loss": 0.0982, + "num_input_tokens_seen": 18256800, + "step": 14075 + }, + { + "epoch": 0.6879534849632326, + "grad_norm": 0.0014080044347792864, + "learning_rate": 0.2172680125272485, + "loss": 0.1425, + "num_input_tokens_seen": 18262656, + "step": 14080 + }, + { + "epoch": 0.6881977866269269, + "grad_norm": 0.0017105932347476482, + "learning_rate": 0.2172153578104002, + "loss": 0.079, + "num_input_tokens_seen": 18269216, + "step": 14085 + }, + { + "epoch": 0.6884420882906213, + "grad_norm": 0.0011262798216193914, + "learning_rate": 0.21716269272809902, + "loss": 0.0994, + "num_input_tokens_seen": 18275808, + "step": 14090 + }, + { + "epoch": 0.6886863899543156, + "grad_norm": 0.0021661343052983284, + "learning_rate": 0.21711001728846666, + "loss": 0.1241, + "num_input_tokens_seen": 18282432, + "step": 14095 + }, + { + "epoch": 0.6889306916180099, + "grad_norm": 0.0011584728490561247, + "learning_rate": 0.21705733149962628, + "loss": 0.0976, + "num_input_tokens_seen": 18288672, + "step": 14100 + }, + { + "epoch": 0.6891749932817043, + "grad_norm": 0.0013083278900012374, + "learning_rate": 0.21700463536970263, + "loss": 0.1004, + "num_input_tokens_seen": 18295360, + "step": 14105 + }, + { + "epoch": 0.6894192949453986, + "grad_norm": 0.0011242504697293043, + "learning_rate": 0.21695192890682222, + "loss": 0.1103, + "num_input_tokens_seen": 18301728, + "step": 14110 + }, + { + "epoch": 0.6896635966090929, + "grad_norm": 0.00263860821723938, + "learning_rate": 0.21689921211911298, + "loss": 0.0944, + "num_input_tokens_seen": 18308544, + "step": 14115 + }, + { + "epoch": 0.6899078982727872, + "grad_norm": 0.0012811344349756837, + "learning_rate": 0.21684648501470452, + "loss": 0.1215, + "num_input_tokens_seen": 18314688, + "step": 14120 + }, + { + "epoch": 0.6901521999364816, + "grad_norm": 0.0016054412117227912, + "learning_rate": 0.216793747601728, + "loss": 0.115, + "num_input_tokens_seen": 18321312, + "step": 14125 + }, + { + "epoch": 0.6903965016001758, + "grad_norm": 0.0012331443140283227, + "learning_rate": 0.21674099988831627, + "loss": 0.1052, + "num_input_tokens_seen": 18327872, + "step": 14130 + }, + { + "epoch": 0.6906408032638702, + "grad_norm": 0.0014157916884869337, + "learning_rate": 0.21668824188260363, + "loss": 0.1123, + "num_input_tokens_seen": 18334848, + "step": 14135 + }, + { + "epoch": 0.6908851049275646, + "grad_norm": 0.001494238036684692, + "learning_rate": 0.21663547359272606, + "loss": 0.1072, + "num_input_tokens_seen": 18341184, + "step": 14140 + }, + { + "epoch": 0.6911294065912589, + "grad_norm": 0.0008628632058389485, + "learning_rate": 0.216582695026821, + "loss": 0.1006, + "num_input_tokens_seen": 18348224, + "step": 14145 + }, + { + "epoch": 0.6913737082549533, + "grad_norm": 0.0011070476612076163, + "learning_rate": 0.21652990619302767, + "loss": 0.0961, + "num_input_tokens_seen": 18354944, + "step": 14150 + }, + { + "epoch": 0.6916180099186475, + "grad_norm": 0.0013004568172618747, + "learning_rate": 0.21647710709948673, + "loss": 0.1233, + "num_input_tokens_seen": 18361248, + "step": 14155 + }, + { + "epoch": 0.6918623115823419, + "grad_norm": 0.001267589395865798, + "learning_rate": 0.2164242977543405, + "loss": 0.1069, + "num_input_tokens_seen": 18367488, + "step": 14160 + }, + { + "epoch": 0.6921066132460362, + "grad_norm": 0.001974388724192977, + "learning_rate": 0.21637147816573277, + "loss": 0.1148, + "num_input_tokens_seen": 18373600, + "step": 14165 + }, + { + "epoch": 0.6923509149097306, + "grad_norm": 0.0014412750024348497, + "learning_rate": 0.21631864834180908, + "loss": 0.0923, + "num_input_tokens_seen": 18380128, + "step": 14170 + }, + { + "epoch": 0.6925952165734248, + "grad_norm": 0.0012480131117627025, + "learning_rate": 0.21626580829071637, + "loss": 0.1085, + "num_input_tokens_seen": 18386816, + "step": 14175 + }, + { + "epoch": 0.6928395182371192, + "grad_norm": 0.0013769628712907434, + "learning_rate": 0.21621295802060328, + "loss": 0.1153, + "num_input_tokens_seen": 18393120, + "step": 14180 + }, + { + "epoch": 0.6930838199008135, + "grad_norm": 0.001106621464714408, + "learning_rate": 0.21616009753961996, + "loss": 0.0871, + "num_input_tokens_seen": 18399648, + "step": 14185 + }, + { + "epoch": 0.6933281215645078, + "grad_norm": 0.001167361275292933, + "learning_rate": 0.2161072268559182, + "loss": 0.1095, + "num_input_tokens_seen": 18405856, + "step": 14190 + }, + { + "epoch": 0.6935724232282022, + "grad_norm": 0.001347559504210949, + "learning_rate": 0.21605434597765133, + "loss": 0.1098, + "num_input_tokens_seen": 18411936, + "step": 14195 + }, + { + "epoch": 0.6938167248918965, + "grad_norm": 0.0008006177376955748, + "learning_rate": 0.21600145491297418, + "loss": 0.1218, + "num_input_tokens_seen": 18418528, + "step": 14200 + }, + { + "epoch": 0.6938167248918965, + "eval_loss": 0.10513930022716522, + "eval_runtime": 402.7018, + "eval_samples_per_second": 90.352, + "eval_steps_per_second": 22.59, + "num_input_tokens_seen": 18418528, + "step": 14200 + }, + { + "epoch": 0.6940610265555909, + "grad_norm": 0.0013715012464672327, + "learning_rate": 0.21594855367004326, + "loss": 0.1158, + "num_input_tokens_seen": 18424896, + "step": 14205 + }, + { + "epoch": 0.6943053282192851, + "grad_norm": 0.0006306126015260816, + "learning_rate": 0.21589564225701663, + "loss": 0.0819, + "num_input_tokens_seen": 18431104, + "step": 14210 + }, + { + "epoch": 0.6945496298829795, + "grad_norm": 0.0012065808987244964, + "learning_rate": 0.21584272068205385, + "loss": 0.0912, + "num_input_tokens_seen": 18437248, + "step": 14215 + }, + { + "epoch": 0.6947939315466738, + "grad_norm": 0.0014711193507537246, + "learning_rate": 0.2157897889533161, + "loss": 0.1273, + "num_input_tokens_seen": 18443680, + "step": 14220 + }, + { + "epoch": 0.6950382332103682, + "grad_norm": 0.001476883189752698, + "learning_rate": 0.21573684707896612, + "loss": 0.1078, + "num_input_tokens_seen": 18449728, + "step": 14225 + }, + { + "epoch": 0.6952825348740624, + "grad_norm": 0.001298818620853126, + "learning_rate": 0.21568389506716826, + "loss": 0.1098, + "num_input_tokens_seen": 18456000, + "step": 14230 + }, + { + "epoch": 0.6955268365377568, + "grad_norm": 0.0009593939175829291, + "learning_rate": 0.21563093292608831, + "loss": 0.0882, + "num_input_tokens_seen": 18462784, + "step": 14235 + }, + { + "epoch": 0.6957711382014512, + "grad_norm": 0.0018524278420954943, + "learning_rate": 0.21557796066389376, + "loss": 0.1109, + "num_input_tokens_seen": 18469376, + "step": 14240 + }, + { + "epoch": 0.6960154398651455, + "grad_norm": 0.0015013031661510468, + "learning_rate": 0.21552497828875353, + "loss": 0.1188, + "num_input_tokens_seen": 18475872, + "step": 14245 + }, + { + "epoch": 0.6962597415288398, + "grad_norm": 0.001084623159840703, + "learning_rate": 0.21547198580883828, + "loss": 0.1348, + "num_input_tokens_seen": 18482944, + "step": 14250 + }, + { + "epoch": 0.6965040431925341, + "grad_norm": 0.0010323261376470327, + "learning_rate": 0.21541898323232, + "loss": 0.1011, + "num_input_tokens_seen": 18489568, + "step": 14255 + }, + { + "epoch": 0.6967483448562285, + "grad_norm": 0.0007526400149799883, + "learning_rate": 0.2153659705673724, + "loss": 0.0882, + "num_input_tokens_seen": 18495904, + "step": 14260 + }, + { + "epoch": 0.6969926465199228, + "grad_norm": 0.001664018607698381, + "learning_rate": 0.2153129478221707, + "loss": 0.1105, + "num_input_tokens_seen": 18502784, + "step": 14265 + }, + { + "epoch": 0.6972369481836171, + "grad_norm": 0.000916960765607655, + "learning_rate": 0.21525991500489164, + "loss": 0.0802, + "num_input_tokens_seen": 18509376, + "step": 14270 + }, + { + "epoch": 0.6974812498473114, + "grad_norm": 0.001735823811031878, + "learning_rate": 0.21520687212371362, + "loss": 0.0951, + "num_input_tokens_seen": 18515648, + "step": 14275 + }, + { + "epoch": 0.6977255515110058, + "grad_norm": 0.0014422232052311301, + "learning_rate": 0.21515381918681648, + "loss": 0.099, + "num_input_tokens_seen": 18522272, + "step": 14280 + }, + { + "epoch": 0.6979698531747002, + "grad_norm": 0.0011515446240082383, + "learning_rate": 0.21510075620238167, + "loss": 0.0599, + "num_input_tokens_seen": 18528544, + "step": 14285 + }, + { + "epoch": 0.6982141548383944, + "grad_norm": 0.0018398716347292066, + "learning_rate": 0.21504768317859208, + "loss": 0.0998, + "num_input_tokens_seen": 18535104, + "step": 14290 + }, + { + "epoch": 0.6984584565020888, + "grad_norm": 0.0017990607302635908, + "learning_rate": 0.2149946001236323, + "loss": 0.1368, + "num_input_tokens_seen": 18541632, + "step": 14295 + }, + { + "epoch": 0.6987027581657831, + "grad_norm": 0.0012015000684186816, + "learning_rate": 0.21494150704568848, + "loss": 0.0835, + "num_input_tokens_seen": 18548320, + "step": 14300 + }, + { + "epoch": 0.6989470598294775, + "grad_norm": 0.0006091490504331887, + "learning_rate": 0.21488840395294811, + "loss": 0.1024, + "num_input_tokens_seen": 18554912, + "step": 14305 + }, + { + "epoch": 0.6991913614931717, + "grad_norm": 0.00191246357280761, + "learning_rate": 0.21483529085360042, + "loss": 0.096, + "num_input_tokens_seen": 18562688, + "step": 14310 + }, + { + "epoch": 0.6994356631568661, + "grad_norm": 0.0008834130130708218, + "learning_rate": 0.2147821677558361, + "loss": 0.0933, + "num_input_tokens_seen": 18569568, + "step": 14315 + }, + { + "epoch": 0.6996799648205604, + "grad_norm": 0.000996214454062283, + "learning_rate": 0.2147290346678475, + "loss": 0.1243, + "num_input_tokens_seen": 18575616, + "step": 14320 + }, + { + "epoch": 0.6999242664842548, + "grad_norm": 0.0010322585003450513, + "learning_rate": 0.21467589159782827, + "loss": 0.1134, + "num_input_tokens_seen": 18581696, + "step": 14325 + }, + { + "epoch": 0.700168568147949, + "grad_norm": 0.001561429351568222, + "learning_rate": 0.21462273855397374, + "loss": 0.1105, + "num_input_tokens_seen": 18588160, + "step": 14330 + }, + { + "epoch": 0.7004128698116434, + "grad_norm": 0.002537148306146264, + "learning_rate": 0.21456957554448083, + "loss": 0.105, + "num_input_tokens_seen": 18594624, + "step": 14335 + }, + { + "epoch": 0.7006571714753378, + "grad_norm": 0.0011914814822375774, + "learning_rate": 0.21451640257754795, + "loss": 0.1017, + "num_input_tokens_seen": 18600928, + "step": 14340 + }, + { + "epoch": 0.7009014731390321, + "grad_norm": 0.0010908221593126655, + "learning_rate": 0.21446321966137508, + "loss": 0.115, + "num_input_tokens_seen": 18607488, + "step": 14345 + }, + { + "epoch": 0.7011457748027264, + "grad_norm": 0.0010747071355581284, + "learning_rate": 0.21441002680416354, + "loss": 0.0887, + "num_input_tokens_seen": 18614432, + "step": 14350 + }, + { + "epoch": 0.7013900764664207, + "grad_norm": 0.0008818146889097989, + "learning_rate": 0.21435682401411654, + "loss": 0.1194, + "num_input_tokens_seen": 18620160, + "step": 14355 + }, + { + "epoch": 0.7016343781301151, + "grad_norm": 0.0010225147707387805, + "learning_rate": 0.2143036112994385, + "loss": 0.0945, + "num_input_tokens_seen": 18626656, + "step": 14360 + }, + { + "epoch": 0.7018786797938094, + "grad_norm": 0.0006813909858465195, + "learning_rate": 0.21425038866833548, + "loss": 0.0884, + "num_input_tokens_seen": 18633152, + "step": 14365 + }, + { + "epoch": 0.7021229814575037, + "grad_norm": 0.0011509148171171546, + "learning_rate": 0.21419715612901508, + "loss": 0.1191, + "num_input_tokens_seen": 18639936, + "step": 14370 + }, + { + "epoch": 0.702367283121198, + "grad_norm": 0.0011827992275357246, + "learning_rate": 0.21414391368968652, + "loss": 0.0999, + "num_input_tokens_seen": 18646528, + "step": 14375 + }, + { + "epoch": 0.7026115847848924, + "grad_norm": 0.0016826624050736427, + "learning_rate": 0.21409066135856034, + "loss": 0.0848, + "num_input_tokens_seen": 18653184, + "step": 14380 + }, + { + "epoch": 0.7028558864485868, + "grad_norm": 0.0011423801770433784, + "learning_rate": 0.21403739914384878, + "loss": 0.0959, + "num_input_tokens_seen": 18659584, + "step": 14385 + }, + { + "epoch": 0.703100188112281, + "grad_norm": 0.0007517216727137566, + "learning_rate": 0.21398412705376554, + "loss": 0.0819, + "num_input_tokens_seen": 18666368, + "step": 14390 + }, + { + "epoch": 0.7033444897759754, + "grad_norm": 0.002010585740208626, + "learning_rate": 0.2139308450965258, + "loss": 0.1107, + "num_input_tokens_seen": 18672928, + "step": 14395 + }, + { + "epoch": 0.7035887914396697, + "grad_norm": 0.002444559708237648, + "learning_rate": 0.21387755328034638, + "loss": 0.1068, + "num_input_tokens_seen": 18679264, + "step": 14400 + }, + { + "epoch": 0.7035887914396697, + "eval_loss": 0.12211251258850098, + "eval_runtime": 402.3959, + "eval_samples_per_second": 90.421, + "eval_steps_per_second": 22.607, + "num_input_tokens_seen": 18679264, + "step": 14400 + }, + { + "epoch": 0.7038330931033641, + "grad_norm": 0.0009307067957706749, + "learning_rate": 0.2138242516134455, + "loss": 0.1023, + "num_input_tokens_seen": 18685664, + "step": 14405 + }, + { + "epoch": 0.7040773947670583, + "grad_norm": 0.0013404275523498654, + "learning_rate": 0.2137709401040429, + "loss": 0.088, + "num_input_tokens_seen": 18691968, + "step": 14410 + }, + { + "epoch": 0.7043216964307527, + "grad_norm": 0.001957119442522526, + "learning_rate": 0.21371761876036, + "loss": 0.134, + "num_input_tokens_seen": 18698432, + "step": 14415 + }, + { + "epoch": 0.704565998094447, + "grad_norm": 0.0008412471506744623, + "learning_rate": 0.21366428759061956, + "loss": 0.1152, + "num_input_tokens_seen": 18704832, + "step": 14420 + }, + { + "epoch": 0.7048102997581414, + "grad_norm": 0.000977023970335722, + "learning_rate": 0.2136109466030459, + "loss": 0.1379, + "num_input_tokens_seen": 18711392, + "step": 14425 + }, + { + "epoch": 0.7050546014218357, + "grad_norm": 0.0011199304135516286, + "learning_rate": 0.2135575958058649, + "loss": 0.1136, + "num_input_tokens_seen": 18717824, + "step": 14430 + }, + { + "epoch": 0.70529890308553, + "grad_norm": 0.0007922986405901611, + "learning_rate": 0.2135042352073039, + "loss": 0.091, + "num_input_tokens_seen": 18724576, + "step": 14435 + }, + { + "epoch": 0.7055432047492244, + "grad_norm": 0.0013993013417348266, + "learning_rate": 0.2134508648155918, + "loss": 0.1296, + "num_input_tokens_seen": 18731168, + "step": 14440 + }, + { + "epoch": 0.7057875064129187, + "grad_norm": 0.0007323367171920836, + "learning_rate": 0.213397484638959, + "loss": 0.1215, + "num_input_tokens_seen": 18737664, + "step": 14445 + }, + { + "epoch": 0.706031808076613, + "grad_norm": 0.0011453056940808892, + "learning_rate": 0.21334409468563728, + "loss": 0.1135, + "num_input_tokens_seen": 18744000, + "step": 14450 + }, + { + "epoch": 0.7062761097403073, + "grad_norm": 0.0008280187612399459, + "learning_rate": 0.2132906949638602, + "loss": 0.1074, + "num_input_tokens_seen": 18750336, + "step": 14455 + }, + { + "epoch": 0.7065204114040017, + "grad_norm": 0.0014663306064903736, + "learning_rate": 0.21323728548186255, + "loss": 0.1153, + "num_input_tokens_seen": 18757248, + "step": 14460 + }, + { + "epoch": 0.706764713067696, + "grad_norm": 0.001759974518790841, + "learning_rate": 0.21318386624788088, + "loss": 0.1101, + "num_input_tokens_seen": 18763648, + "step": 14465 + }, + { + "epoch": 0.7070090147313903, + "grad_norm": 0.0009783951099961996, + "learning_rate": 0.21313043727015288, + "loss": 0.1029, + "num_input_tokens_seen": 18770560, + "step": 14470 + }, + { + "epoch": 0.7072533163950846, + "grad_norm": 0.0006393740768544376, + "learning_rate": 0.2130769985569182, + "loss": 0.0995, + "num_input_tokens_seen": 18776864, + "step": 14475 + }, + { + "epoch": 0.707497618058779, + "grad_norm": 0.0016947342082858086, + "learning_rate": 0.21302355011641766, + "loss": 0.1165, + "num_input_tokens_seen": 18783968, + "step": 14480 + }, + { + "epoch": 0.7077419197224734, + "grad_norm": 0.0020624424796551466, + "learning_rate": 0.21297009195689365, + "loss": 0.1173, + "num_input_tokens_seen": 18790368, + "step": 14485 + }, + { + "epoch": 0.7079862213861676, + "grad_norm": 0.0011354940943419933, + "learning_rate": 0.21291662408659015, + "loss": 0.1011, + "num_input_tokens_seen": 18796896, + "step": 14490 + }, + { + "epoch": 0.708230523049862, + "grad_norm": 0.0008787659462541342, + "learning_rate": 0.21286314651375254, + "loss": 0.1003, + "num_input_tokens_seen": 18802976, + "step": 14495 + }, + { + "epoch": 0.7084748247135563, + "grad_norm": 0.001676251064054668, + "learning_rate": 0.2128096592466278, + "loss": 0.1073, + "num_input_tokens_seen": 18809536, + "step": 14500 + }, + { + "epoch": 0.7087191263772507, + "grad_norm": 0.0018803216516971588, + "learning_rate": 0.21275616229346428, + "loss": 0.0953, + "num_input_tokens_seen": 18816000, + "step": 14505 + }, + { + "epoch": 0.7089634280409449, + "grad_norm": 0.0011570851784199476, + "learning_rate": 0.21270265566251184, + "loss": 0.0937, + "num_input_tokens_seen": 18822432, + "step": 14510 + }, + { + "epoch": 0.7092077297046393, + "grad_norm": 0.0012697946513071656, + "learning_rate": 0.21264913936202193, + "loss": 0.1128, + "num_input_tokens_seen": 18828864, + "step": 14515 + }, + { + "epoch": 0.7094520313683336, + "grad_norm": 0.0010867499513551593, + "learning_rate": 0.2125956134002475, + "loss": 0.0916, + "num_input_tokens_seen": 18835744, + "step": 14520 + }, + { + "epoch": 0.709696333032028, + "grad_norm": 0.0013414350105449557, + "learning_rate": 0.2125420777854428, + "loss": 0.0899, + "num_input_tokens_seen": 18842336, + "step": 14525 + }, + { + "epoch": 0.7099406346957223, + "grad_norm": 0.0010557138593867421, + "learning_rate": 0.21248853252586372, + "loss": 0.1067, + "num_input_tokens_seen": 18848736, + "step": 14530 + }, + { + "epoch": 0.7101849363594166, + "grad_norm": 0.0018128586234524846, + "learning_rate": 0.21243497762976774, + "loss": 0.142, + "num_input_tokens_seen": 18855136, + "step": 14535 + }, + { + "epoch": 0.710429238023111, + "grad_norm": 0.0015124678611755371, + "learning_rate": 0.21238141310541356, + "loss": 0.1029, + "num_input_tokens_seen": 18861664, + "step": 14540 + }, + { + "epoch": 0.7106735396868052, + "grad_norm": 0.0011416621273383498, + "learning_rate": 0.21232783896106153, + "loss": 0.1096, + "num_input_tokens_seen": 18868000, + "step": 14545 + }, + { + "epoch": 0.7109178413504996, + "grad_norm": 0.0014183786697685719, + "learning_rate": 0.21227425520497345, + "loss": 0.1084, + "num_input_tokens_seen": 18874496, + "step": 14550 + }, + { + "epoch": 0.7111621430141939, + "grad_norm": 0.001431922661140561, + "learning_rate": 0.2122206618454127, + "loss": 0.1271, + "num_input_tokens_seen": 18880864, + "step": 14555 + }, + { + "epoch": 0.7114064446778883, + "grad_norm": 0.0013566809939220548, + "learning_rate": 0.2121670588906439, + "loss": 0.1062, + "num_input_tokens_seen": 18887584, + "step": 14560 + }, + { + "epoch": 0.7116507463415825, + "grad_norm": 0.0017637658165767789, + "learning_rate": 0.21211344634893345, + "loss": 0.0969, + "num_input_tokens_seen": 18894272, + "step": 14565 + }, + { + "epoch": 0.7118950480052769, + "grad_norm": 0.002110852627083659, + "learning_rate": 0.21205982422854897, + "loss": 0.1026, + "num_input_tokens_seen": 18900768, + "step": 14570 + }, + { + "epoch": 0.7121393496689713, + "grad_norm": 0.0016716130776330829, + "learning_rate": 0.21200619253775974, + "loss": 0.1211, + "num_input_tokens_seen": 18907520, + "step": 14575 + }, + { + "epoch": 0.7123836513326656, + "grad_norm": 0.0010580731322988868, + "learning_rate": 0.21195255128483637, + "loss": 0.103, + "num_input_tokens_seen": 18914240, + "step": 14580 + }, + { + "epoch": 0.71262795299636, + "grad_norm": 0.0016082939691841602, + "learning_rate": 0.21189890047805102, + "loss": 0.092, + "num_input_tokens_seen": 18920576, + "step": 14585 + }, + { + "epoch": 0.7128722546600542, + "grad_norm": 0.0013080766657367349, + "learning_rate": 0.21184524012567735, + "loss": 0.1053, + "num_input_tokens_seen": 18926816, + "step": 14590 + }, + { + "epoch": 0.7131165563237486, + "grad_norm": 0.0019165573175996542, + "learning_rate": 0.2117915702359905, + "loss": 0.0997, + "num_input_tokens_seen": 18933376, + "step": 14595 + }, + { + "epoch": 0.7133608579874429, + "grad_norm": 0.0011029442539438605, + "learning_rate": 0.211737890817267, + "loss": 0.0971, + "num_input_tokens_seen": 18940320, + "step": 14600 + }, + { + "epoch": 0.7133608579874429, + "eval_loss": 0.10460801422595978, + "eval_runtime": 401.9541, + "eval_samples_per_second": 90.52, + "eval_steps_per_second": 22.632, + "num_input_tokens_seen": 18940320, + "step": 14600 + }, + { + "epoch": 0.7136051596511372, + "grad_norm": 0.001093795639462769, + "learning_rate": 0.21168420187778483, + "loss": 0.1007, + "num_input_tokens_seen": 18946816, + "step": 14605 + }, + { + "epoch": 0.7138494613148315, + "grad_norm": 0.000767840479966253, + "learning_rate": 0.21163050342582362, + "loss": 0.0974, + "num_input_tokens_seen": 18953408, + "step": 14610 + }, + { + "epoch": 0.7140937629785259, + "grad_norm": 0.0014315928565338254, + "learning_rate": 0.21157679546966426, + "loss": 0.1043, + "num_input_tokens_seen": 18959872, + "step": 14615 + }, + { + "epoch": 0.7143380646422202, + "grad_norm": 0.0010601291432976723, + "learning_rate": 0.2115230780175892, + "loss": 0.1123, + "num_input_tokens_seen": 18966624, + "step": 14620 + }, + { + "epoch": 0.7145823663059145, + "grad_norm": 0.0015698332572355866, + "learning_rate": 0.21146935107788237, + "loss": 0.1465, + "num_input_tokens_seen": 18972768, + "step": 14625 + }, + { + "epoch": 0.7148266679696089, + "grad_norm": 0.0008204998448491096, + "learning_rate": 0.21141561465882916, + "loss": 0.082, + "num_input_tokens_seen": 18979040, + "step": 14630 + }, + { + "epoch": 0.7150709696333032, + "grad_norm": 0.0015324545092880726, + "learning_rate": 0.21136186876871635, + "loss": 0.1331, + "num_input_tokens_seen": 18985408, + "step": 14635 + }, + { + "epoch": 0.7153152712969976, + "grad_norm": 0.0010471346322447062, + "learning_rate": 0.21130811341583225, + "loss": 0.1185, + "num_input_tokens_seen": 18991872, + "step": 14640 + }, + { + "epoch": 0.7155595729606918, + "grad_norm": 0.0007494007586501539, + "learning_rate": 0.21125434860846667, + "loss": 0.0815, + "num_input_tokens_seen": 18998208, + "step": 14645 + }, + { + "epoch": 0.7158038746243862, + "grad_norm": 0.0005579462740570307, + "learning_rate": 0.2112005743549107, + "loss": 0.1271, + "num_input_tokens_seen": 19005024, + "step": 14650 + }, + { + "epoch": 0.7160481762880805, + "grad_norm": 0.00124653079546988, + "learning_rate": 0.21114679066345707, + "loss": 0.1301, + "num_input_tokens_seen": 19011296, + "step": 14655 + }, + { + "epoch": 0.7162924779517749, + "grad_norm": 0.0007289143977686763, + "learning_rate": 0.21109299754239993, + "loss": 0.115, + "num_input_tokens_seen": 19017888, + "step": 14660 + }, + { + "epoch": 0.7165367796154691, + "grad_norm": 0.0010107411071658134, + "learning_rate": 0.21103919500003482, + "loss": 0.1341, + "num_input_tokens_seen": 19024096, + "step": 14665 + }, + { + "epoch": 0.7167810812791635, + "grad_norm": 0.0012530498206615448, + "learning_rate": 0.21098538304465872, + "loss": 0.1077, + "num_input_tokens_seen": 19030848, + "step": 14670 + }, + { + "epoch": 0.7170253829428579, + "grad_norm": 0.0007731114747002721, + "learning_rate": 0.2109315616845702, + "loss": 0.097, + "num_input_tokens_seen": 19037312, + "step": 14675 + }, + { + "epoch": 0.7172696846065522, + "grad_norm": 0.0009509710362181067, + "learning_rate": 0.21087773092806925, + "loss": 0.1145, + "num_input_tokens_seen": 19043360, + "step": 14680 + }, + { + "epoch": 0.7175139862702465, + "grad_norm": 0.0013710715575143695, + "learning_rate": 0.21082389078345704, + "loss": 0.1369, + "num_input_tokens_seen": 19050240, + "step": 14685 + }, + { + "epoch": 0.7177582879339408, + "grad_norm": 0.0008431289461441338, + "learning_rate": 0.2107700412590365, + "loss": 0.1007, + "num_input_tokens_seen": 19056544, + "step": 14690 + }, + { + "epoch": 0.7180025895976352, + "grad_norm": 0.0009769595926627517, + "learning_rate": 0.210716182363112, + "loss": 0.1209, + "num_input_tokens_seen": 19062528, + "step": 14695 + }, + { + "epoch": 0.7182468912613295, + "grad_norm": 0.0013571222079917789, + "learning_rate": 0.2106623141039891, + "loss": 0.0962, + "num_input_tokens_seen": 19068800, + "step": 14700 + }, + { + "epoch": 0.7184911929250238, + "grad_norm": 0.0007787301437929273, + "learning_rate": 0.21060843648997507, + "loss": 0.1057, + "num_input_tokens_seen": 19075296, + "step": 14705 + }, + { + "epoch": 0.7187354945887181, + "grad_norm": 0.0007904932135716081, + "learning_rate": 0.21055454952937844, + "loss": 0.115, + "num_input_tokens_seen": 19081536, + "step": 14710 + }, + { + "epoch": 0.7189797962524125, + "grad_norm": 0.0009283885010518134, + "learning_rate": 0.21050065323050937, + "loss": 0.1213, + "num_input_tokens_seen": 19087904, + "step": 14715 + }, + { + "epoch": 0.7192240979161069, + "grad_norm": 0.0010478958720341325, + "learning_rate": 0.21044674760167928, + "loss": 0.108, + "num_input_tokens_seen": 19094592, + "step": 14720 + }, + { + "epoch": 0.7194683995798011, + "grad_norm": 0.0008511136402375996, + "learning_rate": 0.210392832651201, + "loss": 0.0958, + "num_input_tokens_seen": 19100672, + "step": 14725 + }, + { + "epoch": 0.7197127012434955, + "grad_norm": 0.0014529302716255188, + "learning_rate": 0.210338908387389, + "loss": 0.1104, + "num_input_tokens_seen": 19107296, + "step": 14730 + }, + { + "epoch": 0.7199570029071898, + "grad_norm": 0.001544549479149282, + "learning_rate": 0.21028497481855912, + "loss": 0.1256, + "num_input_tokens_seen": 19113088, + "step": 14735 + }, + { + "epoch": 0.7202013045708842, + "grad_norm": 0.001724213594570756, + "learning_rate": 0.21023103195302847, + "loss": 0.1024, + "num_input_tokens_seen": 19119552, + "step": 14740 + }, + { + "epoch": 0.7204456062345784, + "grad_norm": 0.0015140494797378778, + "learning_rate": 0.21017707979911582, + "loss": 0.1393, + "num_input_tokens_seen": 19125856, + "step": 14745 + }, + { + "epoch": 0.7206899078982728, + "grad_norm": 0.0014052148908376694, + "learning_rate": 0.21012311836514122, + "loss": 0.0965, + "num_input_tokens_seen": 19131904, + "step": 14750 + }, + { + "epoch": 0.7209342095619671, + "grad_norm": 0.0016675853403285146, + "learning_rate": 0.21006914765942622, + "loss": 0.1215, + "num_input_tokens_seen": 19138560, + "step": 14755 + }, + { + "epoch": 0.7211785112256615, + "grad_norm": 0.001015078742057085, + "learning_rate": 0.2100151676902938, + "loss": 0.1113, + "num_input_tokens_seen": 19144736, + "step": 14760 + }, + { + "epoch": 0.7214228128893557, + "grad_norm": 0.0011984796728938818, + "learning_rate": 0.2099611784660683, + "loss": 0.1081, + "num_input_tokens_seen": 19151616, + "step": 14765 + }, + { + "epoch": 0.7216671145530501, + "grad_norm": 0.0011075760703533888, + "learning_rate": 0.20990717999507552, + "loss": 0.1343, + "num_input_tokens_seen": 19157696, + "step": 14770 + }, + { + "epoch": 0.7219114162167445, + "grad_norm": 0.0014573953812941909, + "learning_rate": 0.20985317228564276, + "loss": 0.1336, + "num_input_tokens_seen": 19163968, + "step": 14775 + }, + { + "epoch": 0.7221557178804388, + "grad_norm": 0.001229821122251451, + "learning_rate": 0.20979915534609872, + "loss": 0.1218, + "num_input_tokens_seen": 19170080, + "step": 14780 + }, + { + "epoch": 0.7224000195441331, + "grad_norm": 0.0009785708971321583, + "learning_rate": 0.20974512918477342, + "loss": 0.1027, + "num_input_tokens_seen": 19176608, + "step": 14785 + }, + { + "epoch": 0.7226443212078274, + "grad_norm": 0.0006114417337812483, + "learning_rate": 0.2096910938099984, + "loss": 0.0859, + "num_input_tokens_seen": 19183360, + "step": 14790 + }, + { + "epoch": 0.7228886228715218, + "grad_norm": 0.0010869421530514956, + "learning_rate": 0.2096370492301066, + "loss": 0.1068, + "num_input_tokens_seen": 19189728, + "step": 14795 + }, + { + "epoch": 0.723132924535216, + "grad_norm": 0.0016352611128240824, + "learning_rate": 0.2095829954534323, + "loss": 0.0899, + "num_input_tokens_seen": 19196416, + "step": 14800 + }, + { + "epoch": 0.723132924535216, + "eval_loss": 0.10458497703075409, + "eval_runtime": 402.4511, + "eval_samples_per_second": 90.408, + "eval_steps_per_second": 22.604, + "num_input_tokens_seen": 19196416, + "step": 14800 + }, + { + "epoch": 0.7233772261989104, + "grad_norm": 0.0017678036820143461, + "learning_rate": 0.2095289324883114, + "loss": 0.1037, + "num_input_tokens_seen": 19202560, + "step": 14805 + }, + { + "epoch": 0.7236215278626047, + "grad_norm": 0.0008817220805212855, + "learning_rate": 0.20947486034308097, + "loss": 0.0765, + "num_input_tokens_seen": 19209408, + "step": 14810 + }, + { + "epoch": 0.7238658295262991, + "grad_norm": 0.0008922705892473459, + "learning_rate": 0.2094207790260797, + "loss": 0.1191, + "num_input_tokens_seen": 19215840, + "step": 14815 + }, + { + "epoch": 0.7241101311899935, + "grad_norm": 0.000757337489631027, + "learning_rate": 0.20936668854564758, + "loss": 0.1369, + "num_input_tokens_seen": 19222304, + "step": 14820 + }, + { + "epoch": 0.7243544328536877, + "grad_norm": 0.0012943504843860865, + "learning_rate": 0.20931258891012602, + "loss": 0.0898, + "num_input_tokens_seen": 19229088, + "step": 14825 + }, + { + "epoch": 0.7245987345173821, + "grad_norm": 0.0015318474033847451, + "learning_rate": 0.20925848012785792, + "loss": 0.0798, + "num_input_tokens_seen": 19236160, + "step": 14830 + }, + { + "epoch": 0.7248430361810764, + "grad_norm": 0.0012264747638255358, + "learning_rate": 0.20920436220718747, + "loss": 0.1228, + "num_input_tokens_seen": 19242240, + "step": 14835 + }, + { + "epoch": 0.7250873378447708, + "grad_norm": 0.0017099726246669888, + "learning_rate": 0.20915023515646033, + "loss": 0.0885, + "num_input_tokens_seen": 19249184, + "step": 14840 + }, + { + "epoch": 0.725331639508465, + "grad_norm": 0.0013695572270080447, + "learning_rate": 0.20909609898402368, + "loss": 0.1093, + "num_input_tokens_seen": 19255520, + "step": 14845 + }, + { + "epoch": 0.7255759411721594, + "grad_norm": 0.0022889375686645508, + "learning_rate": 0.2090419536982258, + "loss": 0.1243, + "num_input_tokens_seen": 19261536, + "step": 14850 + }, + { + "epoch": 0.7258202428358537, + "grad_norm": 0.000935673713684082, + "learning_rate": 0.2089877993074168, + "loss": 0.0839, + "num_input_tokens_seen": 19267808, + "step": 14855 + }, + { + "epoch": 0.726064544499548, + "grad_norm": 0.001761270104907453, + "learning_rate": 0.20893363581994784, + "loss": 0.1244, + "num_input_tokens_seen": 19273920, + "step": 14860 + }, + { + "epoch": 0.7263088461632423, + "grad_norm": 0.0010468749096617103, + "learning_rate": 0.2088794632441716, + "loss": 0.0926, + "num_input_tokens_seen": 19280416, + "step": 14865 + }, + { + "epoch": 0.7265531478269367, + "grad_norm": 0.001070650527253747, + "learning_rate": 0.20882528158844219, + "loss": 0.1002, + "num_input_tokens_seen": 19286784, + "step": 14870 + }, + { + "epoch": 0.7267974494906311, + "grad_norm": 0.0017042479012161493, + "learning_rate": 0.20877109086111514, + "loss": 0.1116, + "num_input_tokens_seen": 19293536, + "step": 14875 + }, + { + "epoch": 0.7270417511543253, + "grad_norm": 0.0009979017777368426, + "learning_rate": 0.2087168910705473, + "loss": 0.0981, + "num_input_tokens_seen": 19299968, + "step": 14880 + }, + { + "epoch": 0.7272860528180197, + "grad_norm": 0.001860165037214756, + "learning_rate": 0.208662682225097, + "loss": 0.1168, + "num_input_tokens_seen": 19306496, + "step": 14885 + }, + { + "epoch": 0.727530354481714, + "grad_norm": 0.0012635965831577778, + "learning_rate": 0.2086084643331239, + "loss": 0.1081, + "num_input_tokens_seen": 19312928, + "step": 14890 + }, + { + "epoch": 0.7277746561454084, + "grad_norm": 0.001653542509302497, + "learning_rate": 0.20855423740298906, + "loss": 0.1229, + "num_input_tokens_seen": 19319232, + "step": 14895 + }, + { + "epoch": 0.7280189578091026, + "grad_norm": 0.0009201854118146002, + "learning_rate": 0.208500001443055, + "loss": 0.1103, + "num_input_tokens_seen": 19325984, + "step": 14900 + }, + { + "epoch": 0.728263259472797, + "grad_norm": 0.0011425140546634793, + "learning_rate": 0.20844575646168553, + "loss": 0.106, + "num_input_tokens_seen": 19332000, + "step": 14905 + }, + { + "epoch": 0.7285075611364913, + "grad_norm": 0.001690673641860485, + "learning_rate": 0.20839150246724594, + "loss": 0.1207, + "num_input_tokens_seen": 19338240, + "step": 14910 + }, + { + "epoch": 0.7287518628001857, + "grad_norm": 0.0011493810452520847, + "learning_rate": 0.20833723946810287, + "loss": 0.1045, + "num_input_tokens_seen": 19345056, + "step": 14915 + }, + { + "epoch": 0.72899616446388, + "grad_norm": 0.0018789498135447502, + "learning_rate": 0.20828296747262437, + "loss": 0.0971, + "num_input_tokens_seen": 19351744, + "step": 14920 + }, + { + "epoch": 0.7292404661275743, + "grad_norm": 0.0014973839279264212, + "learning_rate": 0.20822868648917986, + "loss": 0.1043, + "num_input_tokens_seen": 19358144, + "step": 14925 + }, + { + "epoch": 0.7294847677912687, + "grad_norm": 0.0011445212876424193, + "learning_rate": 0.20817439652614017, + "loss": 0.0945, + "num_input_tokens_seen": 19364192, + "step": 14930 + }, + { + "epoch": 0.729729069454963, + "grad_norm": 0.0025958882179111242, + "learning_rate": 0.20812009759187744, + "loss": 0.1199, + "num_input_tokens_seen": 19370080, + "step": 14935 + }, + { + "epoch": 0.7299733711186573, + "grad_norm": 0.0013429742539301515, + "learning_rate": 0.2080657896947653, + "loss": 0.1022, + "num_input_tokens_seen": 19376800, + "step": 14940 + }, + { + "epoch": 0.7302176727823516, + "grad_norm": 0.0007973371539264917, + "learning_rate": 0.2080114728431787, + "loss": 0.0884, + "num_input_tokens_seen": 19383136, + "step": 14945 + }, + { + "epoch": 0.730461974446046, + "grad_norm": 0.0012458801502361894, + "learning_rate": 0.20795714704549392, + "loss": 0.084, + "num_input_tokens_seen": 19389440, + "step": 14950 + }, + { + "epoch": 0.7307062761097403, + "grad_norm": 0.0023581411223858595, + "learning_rate": 0.20790281231008875, + "loss": 0.1028, + "num_input_tokens_seen": 19395904, + "step": 14955 + }, + { + "epoch": 0.7309505777734346, + "grad_norm": 0.0015303794061765075, + "learning_rate": 0.20784846864534226, + "loss": 0.0923, + "num_input_tokens_seen": 19402528, + "step": 14960 + }, + { + "epoch": 0.731194879437129, + "grad_norm": 0.0008654179400764406, + "learning_rate": 0.20779411605963496, + "loss": 0.1006, + "num_input_tokens_seen": 19409056, + "step": 14965 + }, + { + "epoch": 0.7314391811008233, + "grad_norm": 0.0008293463615700603, + "learning_rate": 0.2077397545613487, + "loss": 0.1123, + "num_input_tokens_seen": 19415680, + "step": 14970 + }, + { + "epoch": 0.7316834827645177, + "grad_norm": 0.0015141660114750266, + "learning_rate": 0.20768538415886661, + "loss": 0.1292, + "num_input_tokens_seen": 19421984, + "step": 14975 + }, + { + "epoch": 0.7319277844282119, + "grad_norm": 0.0011097764363512397, + "learning_rate": 0.20763100486057343, + "loss": 0.0901, + "num_input_tokens_seen": 19428480, + "step": 14980 + }, + { + "epoch": 0.7321720860919063, + "grad_norm": 0.0011861312668770552, + "learning_rate": 0.20757661667485502, + "loss": 0.1403, + "num_input_tokens_seen": 19435392, + "step": 14985 + }, + { + "epoch": 0.7324163877556006, + "grad_norm": 0.0013581457315012813, + "learning_rate": 0.2075222196100988, + "loss": 0.1248, + "num_input_tokens_seen": 19441696, + "step": 14990 + }, + { + "epoch": 0.732660689419295, + "grad_norm": 0.0009569172980263829, + "learning_rate": 0.20746781367469344, + "loss": 0.1179, + "num_input_tokens_seen": 19448896, + "step": 14995 + }, + { + "epoch": 0.7329049910829892, + "grad_norm": 0.0012400004779919982, + "learning_rate": 0.207413398877029, + "loss": 0.1098, + "num_input_tokens_seen": 19454912, + "step": 15000 + }, + { + "epoch": 0.7329049910829892, + "eval_loss": 0.10610581934452057, + "eval_runtime": 402.7478, + "eval_samples_per_second": 90.342, + "eval_steps_per_second": 22.587, + "num_input_tokens_seen": 19454912, + "step": 15000 + }, + { + "epoch": 0.7331492927466836, + "grad_norm": 0.0010952516458928585, + "learning_rate": 0.20735897522549698, + "loss": 0.0945, + "num_input_tokens_seen": 19461568, + "step": 15005 + }, + { + "epoch": 0.7333935944103779, + "grad_norm": 0.0011293496936559677, + "learning_rate": 0.2073045427284902, + "loss": 0.112, + "num_input_tokens_seen": 19468736, + "step": 15010 + }, + { + "epoch": 0.7336378960740723, + "grad_norm": 0.002216302091255784, + "learning_rate": 0.2072501013944027, + "loss": 0.1061, + "num_input_tokens_seen": 19475168, + "step": 15015 + }, + { + "epoch": 0.7338821977377666, + "grad_norm": 0.0010592241305857897, + "learning_rate": 0.20719565123163017, + "loss": 0.1075, + "num_input_tokens_seen": 19481728, + "step": 15020 + }, + { + "epoch": 0.7341264994014609, + "grad_norm": 0.0013655207585543394, + "learning_rate": 0.20714119224856944, + "loss": 0.1106, + "num_input_tokens_seen": 19488256, + "step": 15025 + }, + { + "epoch": 0.7343708010651553, + "grad_norm": 0.0011336682364344597, + "learning_rate": 0.2070867244536188, + "loss": 0.1151, + "num_input_tokens_seen": 19494528, + "step": 15030 + }, + { + "epoch": 0.7346151027288496, + "grad_norm": 0.001070053898729384, + "learning_rate": 0.20703224785517785, + "loss": 0.1104, + "num_input_tokens_seen": 19501088, + "step": 15035 + }, + { + "epoch": 0.7348594043925439, + "grad_norm": 0.0011083895806223154, + "learning_rate": 0.20697776246164754, + "loss": 0.1252, + "num_input_tokens_seen": 19507424, + "step": 15040 + }, + { + "epoch": 0.7351037060562382, + "grad_norm": 0.0012467886554077268, + "learning_rate": 0.2069232682814303, + "loss": 0.1068, + "num_input_tokens_seen": 19514080, + "step": 15045 + }, + { + "epoch": 0.7353480077199326, + "grad_norm": 0.001165427966043353, + "learning_rate": 0.20686876532292972, + "loss": 0.111, + "num_input_tokens_seen": 19521056, + "step": 15050 + }, + { + "epoch": 0.7355923093836269, + "grad_norm": 0.001540884841233492, + "learning_rate": 0.20681425359455083, + "loss": 0.0971, + "num_input_tokens_seen": 19527584, + "step": 15055 + }, + { + "epoch": 0.7358366110473212, + "grad_norm": 0.0008897901279851794, + "learning_rate": 0.20675973310470008, + "loss": 0.0896, + "num_input_tokens_seen": 19534368, + "step": 15060 + }, + { + "epoch": 0.7360809127110156, + "grad_norm": 0.0015501469606533647, + "learning_rate": 0.2067052038617852, + "loss": 0.1171, + "num_input_tokens_seen": 19540416, + "step": 15065 + }, + { + "epoch": 0.7363252143747099, + "grad_norm": 0.0010176997166126966, + "learning_rate": 0.2066506658742153, + "loss": 0.094, + "num_input_tokens_seen": 19547296, + "step": 15070 + }, + { + "epoch": 0.7365695160384043, + "grad_norm": 0.0011290665715932846, + "learning_rate": 0.20659611915040077, + "loss": 0.1103, + "num_input_tokens_seen": 19553536, + "step": 15075 + }, + { + "epoch": 0.7368138177020985, + "grad_norm": 0.0015697324415668845, + "learning_rate": 0.20654156369875348, + "loss": 0.1032, + "num_input_tokens_seen": 19559648, + "step": 15080 + }, + { + "epoch": 0.7370581193657929, + "grad_norm": 0.001439630170352757, + "learning_rate": 0.20648699952768648, + "loss": 0.1009, + "num_input_tokens_seen": 19565984, + "step": 15085 + }, + { + "epoch": 0.7373024210294872, + "grad_norm": 0.0012976074358448386, + "learning_rate": 0.20643242664561437, + "loss": 0.0984, + "num_input_tokens_seen": 19573024, + "step": 15090 + }, + { + "epoch": 0.7375467226931816, + "grad_norm": 0.0015858369879424572, + "learning_rate": 0.20637784506095277, + "loss": 0.1382, + "num_input_tokens_seen": 19579360, + "step": 15095 + }, + { + "epoch": 0.7377910243568758, + "grad_norm": 0.0011537396349012852, + "learning_rate": 0.20632325478211908, + "loss": 0.0923, + "num_input_tokens_seen": 19585920, + "step": 15100 + }, + { + "epoch": 0.7380353260205702, + "grad_norm": 0.001701922039501369, + "learning_rate": 0.20626865581753165, + "loss": 0.1122, + "num_input_tokens_seen": 19592544, + "step": 15105 + }, + { + "epoch": 0.7382796276842646, + "grad_norm": 0.0018840249394997954, + "learning_rate": 0.2062140481756104, + "loss": 0.1306, + "num_input_tokens_seen": 19598720, + "step": 15110 + }, + { + "epoch": 0.7385239293479589, + "grad_norm": 0.0010503428056836128, + "learning_rate": 0.20615943186477648, + "loss": 0.1039, + "num_input_tokens_seen": 19605184, + "step": 15115 + }, + { + "epoch": 0.7387682310116532, + "grad_norm": 0.001313105458393693, + "learning_rate": 0.20610480689345242, + "loss": 0.0885, + "num_input_tokens_seen": 19611584, + "step": 15120 + }, + { + "epoch": 0.7390125326753475, + "grad_norm": 0.0015382606070488691, + "learning_rate": 0.2060501732700621, + "loss": 0.1411, + "num_input_tokens_seen": 19618048, + "step": 15125 + }, + { + "epoch": 0.7392568343390419, + "grad_norm": 0.0011066454462707043, + "learning_rate": 0.20599553100303067, + "loss": 0.074, + "num_input_tokens_seen": 19624832, + "step": 15130 + }, + { + "epoch": 0.7395011360027361, + "grad_norm": 0.0021549996454268694, + "learning_rate": 0.20594088010078465, + "loss": 0.085, + "num_input_tokens_seen": 19631200, + "step": 15135 + }, + { + "epoch": 0.7397454376664305, + "grad_norm": 0.00093191082123667, + "learning_rate": 0.20588622057175196, + "loss": 0.0889, + "num_input_tokens_seen": 19638464, + "step": 15140 + }, + { + "epoch": 0.7399897393301248, + "grad_norm": 0.0011326312087476254, + "learning_rate": 0.20583155242436177, + "loss": 0.1156, + "num_input_tokens_seen": 19644576, + "step": 15145 + }, + { + "epoch": 0.7402340409938192, + "grad_norm": 0.0015261840308085084, + "learning_rate": 0.20577687566704453, + "loss": 0.1094, + "num_input_tokens_seen": 19651136, + "step": 15150 + }, + { + "epoch": 0.7404783426575134, + "grad_norm": 0.0016539632342755795, + "learning_rate": 0.20572219030823213, + "loss": 0.0933, + "num_input_tokens_seen": 19658112, + "step": 15155 + }, + { + "epoch": 0.7407226443212078, + "grad_norm": 0.0012088430812582374, + "learning_rate": 0.20566749635635775, + "loss": 0.097, + "num_input_tokens_seen": 19664288, + "step": 15160 + }, + { + "epoch": 0.7409669459849022, + "grad_norm": 0.001750956755131483, + "learning_rate": 0.20561279381985587, + "loss": 0.1427, + "num_input_tokens_seen": 19670688, + "step": 15165 + }, + { + "epoch": 0.7412112476485965, + "grad_norm": 0.0011872055474668741, + "learning_rate": 0.2055580827071623, + "loss": 0.1098, + "num_input_tokens_seen": 19676800, + "step": 15170 + }, + { + "epoch": 0.7414555493122909, + "grad_norm": 0.001099477638490498, + "learning_rate": 0.20550336302671418, + "loss": 0.0978, + "num_input_tokens_seen": 19683008, + "step": 15175 + }, + { + "epoch": 0.7416998509759851, + "grad_norm": 0.0015761443646624684, + "learning_rate": 0.20544863478695, + "loss": 0.1082, + "num_input_tokens_seen": 19689696, + "step": 15180 + }, + { + "epoch": 0.7419441526396795, + "grad_norm": 0.0008079453255049884, + "learning_rate": 0.20539389799630953, + "loss": 0.0889, + "num_input_tokens_seen": 19696384, + "step": 15185 + }, + { + "epoch": 0.7421884543033738, + "grad_norm": 0.0007582379621453583, + "learning_rate": 0.20533915266323388, + "loss": 0.1003, + "num_input_tokens_seen": 19702944, + "step": 15190 + }, + { + "epoch": 0.7424327559670681, + "grad_norm": 0.001015720423310995, + "learning_rate": 0.20528439879616542, + "loss": 0.1045, + "num_input_tokens_seen": 19709344, + "step": 15195 + }, + { + "epoch": 0.7426770576307624, + "grad_norm": 0.0015280848601832986, + "learning_rate": 0.20522963640354794, + "loss": 0.0999, + "num_input_tokens_seen": 19715616, + "step": 15200 + }, + { + "epoch": 0.7426770576307624, + "eval_loss": 0.10302026569843292, + "eval_runtime": 402.3381, + "eval_samples_per_second": 90.434, + "eval_steps_per_second": 22.61, + "num_input_tokens_seen": 19715616, + "step": 15200 + }, + { + "epoch": 0.7429213592944568, + "grad_norm": 0.0013731884537264705, + "learning_rate": 0.20517486549382644, + "loss": 0.1111, + "num_input_tokens_seen": 19722208, + "step": 15205 + }, + { + "epoch": 0.7431656609581512, + "grad_norm": 0.0009719849331304431, + "learning_rate": 0.20512008607544735, + "loss": 0.1261, + "num_input_tokens_seen": 19728256, + "step": 15210 + }, + { + "epoch": 0.7434099626218454, + "grad_norm": 0.0017470731399953365, + "learning_rate": 0.20506529815685826, + "loss": 0.1022, + "num_input_tokens_seen": 19734912, + "step": 15215 + }, + { + "epoch": 0.7436542642855398, + "grad_norm": 0.0011398057686164975, + "learning_rate": 0.2050105017465082, + "loss": 0.1085, + "num_input_tokens_seen": 19741568, + "step": 15220 + }, + { + "epoch": 0.7438985659492341, + "grad_norm": 0.0017200250877067447, + "learning_rate": 0.20495569685284754, + "loss": 0.104, + "num_input_tokens_seen": 19748192, + "step": 15225 + }, + { + "epoch": 0.7441428676129285, + "grad_norm": 0.0008758865878917277, + "learning_rate": 0.20490088348432778, + "loss": 0.0804, + "num_input_tokens_seen": 19754848, + "step": 15230 + }, + { + "epoch": 0.7443871692766227, + "grad_norm": 0.001504877582192421, + "learning_rate": 0.2048460616494018, + "loss": 0.0984, + "num_input_tokens_seen": 19761056, + "step": 15235 + }, + { + "epoch": 0.7446314709403171, + "grad_norm": 0.00097639363957569, + "learning_rate": 0.2047912313565239, + "loss": 0.0821, + "num_input_tokens_seen": 19767200, + "step": 15240 + }, + { + "epoch": 0.7448757726040114, + "grad_norm": 0.0007243599975481629, + "learning_rate": 0.20473639261414958, + "loss": 0.0854, + "num_input_tokens_seen": 19774336, + "step": 15245 + }, + { + "epoch": 0.7451200742677058, + "grad_norm": 0.0010984557447955012, + "learning_rate": 0.2046815454307357, + "loss": 0.0899, + "num_input_tokens_seen": 19781056, + "step": 15250 + }, + { + "epoch": 0.7453643759314001, + "grad_norm": 0.001321729039773345, + "learning_rate": 0.20462668981474028, + "loss": 0.0908, + "num_input_tokens_seen": 19787232, + "step": 15255 + }, + { + "epoch": 0.7456086775950944, + "grad_norm": 0.0010416710283607244, + "learning_rate": 0.20457182577462288, + "loss": 0.0869, + "num_input_tokens_seen": 19794112, + "step": 15260 + }, + { + "epoch": 0.7458529792587888, + "grad_norm": 0.001101493602618575, + "learning_rate": 0.2045169533188441, + "loss": 0.1241, + "num_input_tokens_seen": 19800608, + "step": 15265 + }, + { + "epoch": 0.7460972809224831, + "grad_norm": 0.0015297940699383616, + "learning_rate": 0.20446207245586603, + "loss": 0.095, + "num_input_tokens_seen": 19806720, + "step": 15270 + }, + { + "epoch": 0.7463415825861774, + "grad_norm": 0.0017135459929704666, + "learning_rate": 0.20440718319415196, + "loss": 0.1108, + "num_input_tokens_seen": 19812896, + "step": 15275 + }, + { + "epoch": 0.7465858842498717, + "grad_norm": 0.0009217565529979765, + "learning_rate": 0.20435228554216653, + "loss": 0.1025, + "num_input_tokens_seen": 19819680, + "step": 15280 + }, + { + "epoch": 0.7468301859135661, + "grad_norm": 0.0008536595851182938, + "learning_rate": 0.20429737950837565, + "loss": 0.0735, + "num_input_tokens_seen": 19826624, + "step": 15285 + }, + { + "epoch": 0.7470744875772604, + "grad_norm": 0.001851323526352644, + "learning_rate": 0.20424246510124647, + "loss": 0.116, + "num_input_tokens_seen": 19833504, + "step": 15290 + }, + { + "epoch": 0.7473187892409547, + "grad_norm": 0.0013770924415439367, + "learning_rate": 0.20418754232924755, + "loss": 0.0636, + "num_input_tokens_seen": 19840320, + "step": 15295 + }, + { + "epoch": 0.747563090904649, + "grad_norm": 0.0019773971289396286, + "learning_rate": 0.20413261120084863, + "loss": 0.1082, + "num_input_tokens_seen": 19847072, + "step": 15300 + }, + { + "epoch": 0.7478073925683434, + "grad_norm": 0.0017196357948705554, + "learning_rate": 0.2040776717245208, + "loss": 0.1026, + "num_input_tokens_seen": 19853056, + "step": 15305 + }, + { + "epoch": 0.7480516942320378, + "grad_norm": 0.0013905840460211039, + "learning_rate": 0.2040227239087364, + "loss": 0.1063, + "num_input_tokens_seen": 19859680, + "step": 15310 + }, + { + "epoch": 0.748295995895732, + "grad_norm": 0.0009006684413179755, + "learning_rate": 0.20396776776196904, + "loss": 0.1098, + "num_input_tokens_seen": 19865824, + "step": 15315 + }, + { + "epoch": 0.7485402975594264, + "grad_norm": 0.001460922765545547, + "learning_rate": 0.20391280329269373, + "loss": 0.1256, + "num_input_tokens_seen": 19872256, + "step": 15320 + }, + { + "epoch": 0.7487845992231207, + "grad_norm": 0.0010355369886383414, + "learning_rate": 0.20385783050938663, + "loss": 0.1011, + "num_input_tokens_seen": 19878944, + "step": 15325 + }, + { + "epoch": 0.7490289008868151, + "grad_norm": 0.0012462957529351115, + "learning_rate": 0.20380284942052526, + "loss": 0.0972, + "num_input_tokens_seen": 19884992, + "step": 15330 + }, + { + "epoch": 0.7492732025505093, + "grad_norm": 0.001153809018433094, + "learning_rate": 0.2037478600345884, + "loss": 0.1092, + "num_input_tokens_seen": 19891232, + "step": 15335 + }, + { + "epoch": 0.7495175042142037, + "grad_norm": 0.0016265539452433586, + "learning_rate": 0.20369286236005604, + "loss": 0.0914, + "num_input_tokens_seen": 19897568, + "step": 15340 + }, + { + "epoch": 0.749761805877898, + "grad_norm": 0.0010807692306116223, + "learning_rate": 0.20363785640540957, + "loss": 0.1259, + "num_input_tokens_seen": 19903328, + "step": 15345 + }, + { + "epoch": 0.7500061075415924, + "grad_norm": 0.001183045213110745, + "learning_rate": 0.2035828421791316, + "loss": 0.1294, + "num_input_tokens_seen": 19910080, + "step": 15350 + }, + { + "epoch": 0.7502504092052867, + "grad_norm": 0.000765422941185534, + "learning_rate": 0.20352781968970599, + "loss": 0.1055, + "num_input_tokens_seen": 19917088, + "step": 15355 + }, + { + "epoch": 0.750494710868981, + "grad_norm": 0.0010200864635407925, + "learning_rate": 0.2034727889456179, + "loss": 0.1263, + "num_input_tokens_seen": 19923552, + "step": 15360 + }, + { + "epoch": 0.7507390125326754, + "grad_norm": 0.0007784191984683275, + "learning_rate": 0.2034177499553538, + "loss": 0.1014, + "num_input_tokens_seen": 19929824, + "step": 15365 + }, + { + "epoch": 0.7509833141963697, + "grad_norm": 0.0007641958072781563, + "learning_rate": 0.2033627027274014, + "loss": 0.0838, + "num_input_tokens_seen": 19936512, + "step": 15370 + }, + { + "epoch": 0.751227615860064, + "grad_norm": 0.0010714089730754495, + "learning_rate": 0.20330764727024955, + "loss": 0.0914, + "num_input_tokens_seen": 19942816, + "step": 15375 + }, + { + "epoch": 0.7514719175237583, + "grad_norm": 0.0016234178328886628, + "learning_rate": 0.20325258359238868, + "loss": 0.1101, + "num_input_tokens_seen": 19949216, + "step": 15380 + }, + { + "epoch": 0.7517162191874527, + "grad_norm": 0.0012598736211657524, + "learning_rate": 0.20319751170231018, + "loss": 0.1161, + "num_input_tokens_seen": 19956992, + "step": 15385 + }, + { + "epoch": 0.751960520851147, + "grad_norm": 0.0014911673497408628, + "learning_rate": 0.2031424316085068, + "loss": 0.0749, + "num_input_tokens_seen": 19963712, + "step": 15390 + }, + { + "epoch": 0.7522048225148413, + "grad_norm": 0.0011428177822381258, + "learning_rate": 0.20308734331947265, + "loss": 0.0879, + "num_input_tokens_seen": 19970336, + "step": 15395 + }, + { + "epoch": 0.7524491241785357, + "grad_norm": 0.000924473803024739, + "learning_rate": 0.20303224684370305, + "loss": 0.1159, + "num_input_tokens_seen": 19976768, + "step": 15400 + }, + { + "epoch": 0.7524491241785357, + "eval_loss": 0.10564374178647995, + "eval_runtime": 402.8098, + "eval_samples_per_second": 90.328, + "eval_steps_per_second": 22.584, + "num_input_tokens_seen": 19976768, + "step": 15400 + }, + { + "epoch": 0.75269342584223, + "grad_norm": 0.0013527178671211004, + "learning_rate": 0.20297714218969456, + "loss": 0.0739, + "num_input_tokens_seen": 19983008, + "step": 15405 + }, + { + "epoch": 0.7529377275059244, + "grad_norm": 0.0007929868879728019, + "learning_rate": 0.20292202936594497, + "loss": 0.0708, + "num_input_tokens_seen": 19989280, + "step": 15410 + }, + { + "epoch": 0.7531820291696186, + "grad_norm": 0.0008452790789306164, + "learning_rate": 0.2028669083809534, + "loss": 0.0889, + "num_input_tokens_seen": 19996064, + "step": 15415 + }, + { + "epoch": 0.753426330833313, + "grad_norm": 0.001392271602526307, + "learning_rate": 0.20281177924322016, + "loss": 0.0847, + "num_input_tokens_seen": 20002752, + "step": 15420 + }, + { + "epoch": 0.7536706324970073, + "grad_norm": 0.0010645152069628239, + "learning_rate": 0.2027566419612469, + "loss": 0.0609, + "num_input_tokens_seen": 20009664, + "step": 15425 + }, + { + "epoch": 0.7539149341607017, + "grad_norm": 0.003256015945225954, + "learning_rate": 0.20270149654353647, + "loss": 0.1314, + "num_input_tokens_seen": 20015776, + "step": 15430 + }, + { + "epoch": 0.7541592358243959, + "grad_norm": 0.0013972870074212551, + "learning_rate": 0.202646342998593, + "loss": 0.1082, + "num_input_tokens_seen": 20022272, + "step": 15435 + }, + { + "epoch": 0.7544035374880903, + "grad_norm": 0.0014883102849125862, + "learning_rate": 0.20259118133492185, + "loss": 0.0967, + "num_input_tokens_seen": 20028608, + "step": 15440 + }, + { + "epoch": 0.7546478391517846, + "grad_norm": 0.001374737243168056, + "learning_rate": 0.20253601156102966, + "loss": 0.1085, + "num_input_tokens_seen": 20035424, + "step": 15445 + }, + { + "epoch": 0.754892140815479, + "grad_norm": 0.0019484232179820538, + "learning_rate": 0.20248083368542422, + "loss": 0.1333, + "num_input_tokens_seen": 20041632, + "step": 15450 + }, + { + "epoch": 0.7551364424791733, + "grad_norm": 0.0019904046785086393, + "learning_rate": 0.2024256477166147, + "loss": 0.1122, + "num_input_tokens_seen": 20048000, + "step": 15455 + }, + { + "epoch": 0.7553807441428676, + "grad_norm": 0.0009608721593394876, + "learning_rate": 0.2023704536631115, + "loss": 0.1206, + "num_input_tokens_seen": 20054368, + "step": 15460 + }, + { + "epoch": 0.755625045806562, + "grad_norm": 0.0017239426961168647, + "learning_rate": 0.20231525153342625, + "loss": 0.1294, + "num_input_tokens_seen": 20060512, + "step": 15465 + }, + { + "epoch": 0.7558693474702562, + "grad_norm": 0.000581893022172153, + "learning_rate": 0.20226004133607173, + "loss": 0.105, + "num_input_tokens_seen": 20066880, + "step": 15470 + }, + { + "epoch": 0.7561136491339506, + "grad_norm": 0.0010387457441538572, + "learning_rate": 0.20220482307956214, + "loss": 0.1009, + "num_input_tokens_seen": 20073280, + "step": 15475 + }, + { + "epoch": 0.7563579507976449, + "grad_norm": 0.0007473833975382149, + "learning_rate": 0.20214959677241276, + "loss": 0.0827, + "num_input_tokens_seen": 20079616, + "step": 15480 + }, + { + "epoch": 0.7566022524613393, + "grad_norm": 0.0006733008776791394, + "learning_rate": 0.20209436242314022, + "loss": 0.0845, + "num_input_tokens_seen": 20086496, + "step": 15485 + }, + { + "epoch": 0.7568465541250335, + "grad_norm": 0.0007724790484644473, + "learning_rate": 0.2020391200402623, + "loss": 0.0888, + "num_input_tokens_seen": 20092448, + "step": 15490 + }, + { + "epoch": 0.7570908557887279, + "grad_norm": 0.0021390842739492655, + "learning_rate": 0.2019838696322981, + "loss": 0.0918, + "num_input_tokens_seen": 20098816, + "step": 15495 + }, + { + "epoch": 0.7573351574524223, + "grad_norm": 0.001566532882861793, + "learning_rate": 0.20192861120776798, + "loss": 0.09, + "num_input_tokens_seen": 20104800, + "step": 15500 + }, + { + "epoch": 0.7575794591161166, + "grad_norm": 0.0024168051313608885, + "learning_rate": 0.20187334477519345, + "loss": 0.1249, + "num_input_tokens_seen": 20111264, + "step": 15505 + }, + { + "epoch": 0.757823760779811, + "grad_norm": 0.0010339852888137102, + "learning_rate": 0.20181807034309726, + "loss": 0.0906, + "num_input_tokens_seen": 20117824, + "step": 15510 + }, + { + "epoch": 0.7580680624435052, + "grad_norm": 0.0030586246866732836, + "learning_rate": 0.2017627879200034, + "loss": 0.1296, + "num_input_tokens_seen": 20124576, + "step": 15515 + }, + { + "epoch": 0.7583123641071996, + "grad_norm": 0.0013530225260183215, + "learning_rate": 0.2017074975144372, + "loss": 0.0775, + "num_input_tokens_seen": 20130752, + "step": 15520 + }, + { + "epoch": 0.7585566657708939, + "grad_norm": 0.0010334599064663053, + "learning_rate": 0.20165219913492508, + "loss": 0.1381, + "num_input_tokens_seen": 20137088, + "step": 15525 + }, + { + "epoch": 0.7588009674345882, + "grad_norm": 0.000938054290600121, + "learning_rate": 0.20159689278999468, + "loss": 0.0923, + "num_input_tokens_seen": 20143328, + "step": 15530 + }, + { + "epoch": 0.7590452690982825, + "grad_norm": 0.0008568150224164128, + "learning_rate": 0.20154157848817508, + "loss": 0.0799, + "num_input_tokens_seen": 20149440, + "step": 15535 + }, + { + "epoch": 0.7592895707619769, + "grad_norm": 0.0010331214871257544, + "learning_rate": 0.20148625623799632, + "loss": 0.1368, + "num_input_tokens_seen": 20155520, + "step": 15540 + }, + { + "epoch": 0.7595338724256712, + "grad_norm": 0.0012304430129006505, + "learning_rate": 0.20143092604798984, + "loss": 0.1146, + "num_input_tokens_seen": 20162304, + "step": 15545 + }, + { + "epoch": 0.7597781740893655, + "grad_norm": 0.0010189698077738285, + "learning_rate": 0.2013755879266883, + "loss": 0.0858, + "num_input_tokens_seen": 20169408, + "step": 15550 + }, + { + "epoch": 0.7600224757530599, + "grad_norm": 0.0012668604031205177, + "learning_rate": 0.20132024188262543, + "loss": 0.1166, + "num_input_tokens_seen": 20175872, + "step": 15555 + }, + { + "epoch": 0.7602667774167542, + "grad_norm": 0.0005517890676856041, + "learning_rate": 0.2012648879243363, + "loss": 0.097, + "num_input_tokens_seen": 20182080, + "step": 15560 + }, + { + "epoch": 0.7605110790804486, + "grad_norm": 0.0014296566369011998, + "learning_rate": 0.20120952606035725, + "loss": 0.0971, + "num_input_tokens_seen": 20189056, + "step": 15565 + }, + { + "epoch": 0.7607553807441428, + "grad_norm": 0.001002248958684504, + "learning_rate": 0.20115415629922576, + "loss": 0.1012, + "num_input_tokens_seen": 20195392, + "step": 15570 + }, + { + "epoch": 0.7609996824078372, + "grad_norm": 0.0011251391842961311, + "learning_rate": 0.20109877864948048, + "loss": 0.1168, + "num_input_tokens_seen": 20202112, + "step": 15575 + }, + { + "epoch": 0.7612439840715315, + "grad_norm": 0.0013503546360880136, + "learning_rate": 0.20104339311966138, + "loss": 0.0733, + "num_input_tokens_seen": 20208480, + "step": 15580 + }, + { + "epoch": 0.7614882857352259, + "grad_norm": 0.0010572607861831784, + "learning_rate": 0.2009879997183097, + "loss": 0.0922, + "num_input_tokens_seen": 20215616, + "step": 15585 + }, + { + "epoch": 0.7617325873989201, + "grad_norm": 0.001176823629066348, + "learning_rate": 0.20093259845396763, + "loss": 0.0975, + "num_input_tokens_seen": 20222144, + "step": 15590 + }, + { + "epoch": 0.7619768890626145, + "grad_norm": 0.0015798152890056372, + "learning_rate": 0.20087718933517884, + "loss": 0.0831, + "num_input_tokens_seen": 20228192, + "step": 15595 + }, + { + "epoch": 0.7622211907263089, + "grad_norm": 0.0014130634954199195, + "learning_rate": 0.20082177237048807, + "loss": 0.1484, + "num_input_tokens_seen": 20234592, + "step": 15600 + }, + { + "epoch": 0.7622211907263089, + "eval_loss": 0.10523134469985962, + "eval_runtime": 402.3464, + "eval_samples_per_second": 90.432, + "eval_steps_per_second": 22.61, + "num_input_tokens_seen": 20234592, + "step": 15600 + }, + { + "epoch": 0.7624654923900032, + "grad_norm": 0.0012381132692098618, + "learning_rate": 0.20076634756844133, + "loss": 0.0877, + "num_input_tokens_seen": 20241312, + "step": 15605 + }, + { + "epoch": 0.7627097940536975, + "grad_norm": 0.0034974117297679186, + "learning_rate": 0.20071091493758586, + "loss": 0.1588, + "num_input_tokens_seen": 20247648, + "step": 15610 + }, + { + "epoch": 0.7629540957173918, + "grad_norm": 0.0015378015814349055, + "learning_rate": 0.20065547448647003, + "loss": 0.0829, + "num_input_tokens_seen": 20254112, + "step": 15615 + }, + { + "epoch": 0.7631983973810862, + "grad_norm": 0.001190093462355435, + "learning_rate": 0.20060002622364348, + "loss": 0.1186, + "num_input_tokens_seen": 20260864, + "step": 15620 + }, + { + "epoch": 0.7634426990447805, + "grad_norm": 0.0009203033987432718, + "learning_rate": 0.20054457015765695, + "loss": 0.1084, + "num_input_tokens_seen": 20267136, + "step": 15625 + }, + { + "epoch": 0.7636870007084748, + "grad_norm": 0.0008915265207178891, + "learning_rate": 0.20048910629706254, + "loss": 0.0846, + "num_input_tokens_seen": 20273536, + "step": 15630 + }, + { + "epoch": 0.7639313023721691, + "grad_norm": 0.002558245789259672, + "learning_rate": 0.20043363465041347, + "loss": 0.1035, + "num_input_tokens_seen": 20279968, + "step": 15635 + }, + { + "epoch": 0.7641756040358635, + "grad_norm": 0.0008988674380816519, + "learning_rate": 0.2003781552262641, + "loss": 0.1039, + "num_input_tokens_seen": 20286336, + "step": 15640 + }, + { + "epoch": 0.7644199056995579, + "grad_norm": 0.0018469654023647308, + "learning_rate": 0.20032266803317014, + "loss": 0.1043, + "num_input_tokens_seen": 20292448, + "step": 15645 + }, + { + "epoch": 0.7646642073632521, + "grad_norm": 0.0014237507712095976, + "learning_rate": 0.2002671730796884, + "loss": 0.1204, + "num_input_tokens_seen": 20298592, + "step": 15650 + }, + { + "epoch": 0.7649085090269465, + "grad_norm": 0.0007248049951158464, + "learning_rate": 0.20021167037437684, + "loss": 0.0924, + "num_input_tokens_seen": 20305120, + "step": 15655 + }, + { + "epoch": 0.7651528106906408, + "grad_norm": 0.0007246076129376888, + "learning_rate": 0.20015615992579472, + "loss": 0.0773, + "num_input_tokens_seen": 20311424, + "step": 15660 + }, + { + "epoch": 0.7653971123543352, + "grad_norm": 0.0008474981295876205, + "learning_rate": 0.20010064174250244, + "loss": 0.094, + "num_input_tokens_seen": 20317952, + "step": 15665 + }, + { + "epoch": 0.7656414140180294, + "grad_norm": 0.00119930284563452, + "learning_rate": 0.2000451158330616, + "loss": 0.1066, + "num_input_tokens_seen": 20324288, + "step": 15670 + }, + { + "epoch": 0.7658857156817238, + "grad_norm": 0.0010315728140994906, + "learning_rate": 0.199989582206035, + "loss": 0.1118, + "num_input_tokens_seen": 20330496, + "step": 15675 + }, + { + "epoch": 0.7661300173454181, + "grad_norm": 0.001236599637195468, + "learning_rate": 0.1999340408699866, + "loss": 0.097, + "num_input_tokens_seen": 20336608, + "step": 15680 + }, + { + "epoch": 0.7663743190091125, + "grad_norm": 0.002774374093860388, + "learning_rate": 0.19987849183348155, + "loss": 0.0955, + "num_input_tokens_seen": 20344064, + "step": 15685 + }, + { + "epoch": 0.7666186206728067, + "grad_norm": 0.0006983436178416014, + "learning_rate": 0.19982293510508628, + "loss": 0.1271, + "num_input_tokens_seen": 20350432, + "step": 15690 + }, + { + "epoch": 0.7668629223365011, + "grad_norm": 0.0008973755175247788, + "learning_rate": 0.19976737069336833, + "loss": 0.0996, + "num_input_tokens_seen": 20357056, + "step": 15695 + }, + { + "epoch": 0.7671072240001955, + "grad_norm": 0.0011838196078315377, + "learning_rate": 0.1997117986068964, + "loss": 0.0894, + "num_input_tokens_seen": 20363520, + "step": 15700 + }, + { + "epoch": 0.7673515256638898, + "grad_norm": 0.0011125754099339247, + "learning_rate": 0.19965621885424037, + "loss": 0.0818, + "num_input_tokens_seen": 20369856, + "step": 15705 + }, + { + "epoch": 0.7675958273275841, + "grad_norm": 0.000919784652069211, + "learning_rate": 0.19960063144397142, + "loss": 0.1136, + "num_input_tokens_seen": 20376192, + "step": 15710 + }, + { + "epoch": 0.7678401289912784, + "grad_norm": 0.0008339039632119238, + "learning_rate": 0.19954503638466176, + "loss": 0.0852, + "num_input_tokens_seen": 20383360, + "step": 15715 + }, + { + "epoch": 0.7680844306549728, + "grad_norm": 0.000987753737717867, + "learning_rate": 0.1994894336848848, + "loss": 0.0961, + "num_input_tokens_seen": 20390016, + "step": 15720 + }, + { + "epoch": 0.768328732318667, + "grad_norm": 0.0015016074758023024, + "learning_rate": 0.1994338233532153, + "loss": 0.0938, + "num_input_tokens_seen": 20396224, + "step": 15725 + }, + { + "epoch": 0.7685730339823614, + "grad_norm": 0.0008511049672961235, + "learning_rate": 0.19937820539822904, + "loss": 0.0922, + "num_input_tokens_seen": 20402592, + "step": 15730 + }, + { + "epoch": 0.7688173356460557, + "grad_norm": 0.0012548016384243965, + "learning_rate": 0.199322579828503, + "loss": 0.0898, + "num_input_tokens_seen": 20408896, + "step": 15735 + }, + { + "epoch": 0.7690616373097501, + "grad_norm": 0.0009810593910515308, + "learning_rate": 0.19926694665261527, + "loss": 0.0853, + "num_input_tokens_seen": 20415808, + "step": 15740 + }, + { + "epoch": 0.7693059389734445, + "grad_norm": 0.0012905594194307923, + "learning_rate": 0.19921130587914526, + "loss": 0.0859, + "num_input_tokens_seen": 20422336, + "step": 15745 + }, + { + "epoch": 0.7695502406371387, + "grad_norm": 0.0011328160762786865, + "learning_rate": 0.19915565751667344, + "loss": 0.0933, + "num_input_tokens_seen": 20428960, + "step": 15750 + }, + { + "epoch": 0.7697945423008331, + "grad_norm": 0.0011426383862271905, + "learning_rate": 0.19910000157378152, + "loss": 0.1142, + "num_input_tokens_seen": 20435712, + "step": 15755 + }, + { + "epoch": 0.7700388439645274, + "grad_norm": 0.0009603172657079995, + "learning_rate": 0.1990443380590523, + "loss": 0.1598, + "num_input_tokens_seen": 20441856, + "step": 15760 + }, + { + "epoch": 0.7702831456282218, + "grad_norm": 0.0013870041584596038, + "learning_rate": 0.19898866698106984, + "loss": 0.1288, + "num_input_tokens_seen": 20447936, + "step": 15765 + }, + { + "epoch": 0.770527447291916, + "grad_norm": 0.0014180606231093407, + "learning_rate": 0.19893298834841933, + "loss": 0.1035, + "num_input_tokens_seen": 20454432, + "step": 15770 + }, + { + "epoch": 0.7707717489556104, + "grad_norm": 0.000803993723820895, + "learning_rate": 0.19887730216968705, + "loss": 0.0938, + "num_input_tokens_seen": 20460960, + "step": 15775 + }, + { + "epoch": 0.7710160506193047, + "grad_norm": 0.000991548877209425, + "learning_rate": 0.19882160845346053, + "loss": 0.1076, + "num_input_tokens_seen": 20467392, + "step": 15780 + }, + { + "epoch": 0.771260352282999, + "grad_norm": 0.0011664307676255703, + "learning_rate": 0.1987659072083285, + "loss": 0.1039, + "num_input_tokens_seen": 20473856, + "step": 15785 + }, + { + "epoch": 0.7715046539466934, + "grad_norm": 0.0013618008233606815, + "learning_rate": 0.1987101984428807, + "loss": 0.1075, + "num_input_tokens_seen": 20480192, + "step": 15790 + }, + { + "epoch": 0.7717489556103877, + "grad_norm": 0.0010552066378295422, + "learning_rate": 0.19865448216570822, + "loss": 0.0862, + "num_input_tokens_seen": 20486720, + "step": 15795 + }, + { + "epoch": 0.7719932572740821, + "grad_norm": 0.0012246262049302459, + "learning_rate": 0.19859875838540317, + "loss": 0.0815, + "num_input_tokens_seen": 20493056, + "step": 15800 + }, + { + "epoch": 0.7719932572740821, + "eval_loss": 0.10217253118753433, + "eval_runtime": 402.1308, + "eval_samples_per_second": 90.481, + "eval_steps_per_second": 22.622, + "num_input_tokens_seen": 20493056, + "step": 15800 + }, + { + "epoch": 0.7722375589377763, + "grad_norm": 0.001399007742293179, + "learning_rate": 0.1985430271105588, + "loss": 0.105, + "num_input_tokens_seen": 20498880, + "step": 15805 + }, + { + "epoch": 0.7724818606014707, + "grad_norm": 0.0017724192002788186, + "learning_rate": 0.19848728834976961, + "loss": 0.0804, + "num_input_tokens_seen": 20505152, + "step": 15810 + }, + { + "epoch": 0.772726162265165, + "grad_norm": 0.0013883283827453852, + "learning_rate": 0.19843154211163128, + "loss": 0.1109, + "num_input_tokens_seen": 20511232, + "step": 15815 + }, + { + "epoch": 0.7729704639288594, + "grad_norm": 0.0011523272842168808, + "learning_rate": 0.1983757884047405, + "loss": 0.0862, + "num_input_tokens_seen": 20517824, + "step": 15820 + }, + { + "epoch": 0.7732147655925536, + "grad_norm": 0.0011178639251738787, + "learning_rate": 0.1983200272376952, + "loss": 0.0914, + "num_input_tokens_seen": 20524448, + "step": 15825 + }, + { + "epoch": 0.773459067256248, + "grad_norm": 0.0014000667724758387, + "learning_rate": 0.1982642586190945, + "loss": 0.1238, + "num_input_tokens_seen": 20530624, + "step": 15830 + }, + { + "epoch": 0.7737033689199423, + "grad_norm": 0.0017013916512951255, + "learning_rate": 0.1982084825575386, + "loss": 0.1031, + "num_input_tokens_seen": 20537568, + "step": 15835 + }, + { + "epoch": 0.7739476705836367, + "grad_norm": 0.0011306875385344028, + "learning_rate": 0.19815269906162883, + "loss": 0.1237, + "num_input_tokens_seen": 20544032, + "step": 15840 + }, + { + "epoch": 0.774191972247331, + "grad_norm": 0.0015413506189361215, + "learning_rate": 0.19809690813996775, + "loss": 0.1042, + "num_input_tokens_seen": 20550336, + "step": 15845 + }, + { + "epoch": 0.7744362739110253, + "grad_norm": 0.0006359731778502464, + "learning_rate": 0.19804110980115905, + "loss": 0.0963, + "num_input_tokens_seen": 20556832, + "step": 15850 + }, + { + "epoch": 0.7746805755747197, + "grad_norm": 0.0008455241331830621, + "learning_rate": 0.19798530405380746, + "loss": 0.1041, + "num_input_tokens_seen": 20562944, + "step": 15855 + }, + { + "epoch": 0.774924877238414, + "grad_norm": 0.0011427499121055007, + "learning_rate": 0.19792949090651893, + "loss": 0.0951, + "num_input_tokens_seen": 20569888, + "step": 15860 + }, + { + "epoch": 0.7751691789021083, + "grad_norm": 0.0012977448059245944, + "learning_rate": 0.19787367036790066, + "loss": 0.0971, + "num_input_tokens_seen": 20576736, + "step": 15865 + }, + { + "epoch": 0.7754134805658026, + "grad_norm": 0.000978304073214531, + "learning_rate": 0.19781784244656075, + "loss": 0.0817, + "num_input_tokens_seen": 20582752, + "step": 15870 + }, + { + "epoch": 0.775657782229497, + "grad_norm": 0.0011025734711438417, + "learning_rate": 0.19776200715110864, + "loss": 0.0985, + "num_input_tokens_seen": 20589024, + "step": 15875 + }, + { + "epoch": 0.7759020838931913, + "grad_norm": 0.0007806618814356625, + "learning_rate": 0.1977061644901548, + "loss": 0.1018, + "num_input_tokens_seen": 20595520, + "step": 15880 + }, + { + "epoch": 0.7761463855568856, + "grad_norm": 0.001044300850480795, + "learning_rate": 0.1976503144723109, + "loss": 0.0918, + "num_input_tokens_seen": 20601504, + "step": 15885 + }, + { + "epoch": 0.77639068722058, + "grad_norm": 0.0012077587889507413, + "learning_rate": 0.19759445710618967, + "loss": 0.1403, + "num_input_tokens_seen": 20607744, + "step": 15890 + }, + { + "epoch": 0.7766349888842743, + "grad_norm": 0.001149542978964746, + "learning_rate": 0.19753859240040508, + "loss": 0.097, + "num_input_tokens_seen": 20613952, + "step": 15895 + }, + { + "epoch": 0.7768792905479687, + "grad_norm": 0.0016750188078731298, + "learning_rate": 0.1974827203635721, + "loss": 0.1103, + "num_input_tokens_seen": 20620800, + "step": 15900 + }, + { + "epoch": 0.7771235922116629, + "grad_norm": 0.000550419616047293, + "learning_rate": 0.19742684100430694, + "loss": 0.087, + "num_input_tokens_seen": 20627328, + "step": 15905 + }, + { + "epoch": 0.7773678938753573, + "grad_norm": 0.001175620243884623, + "learning_rate": 0.19737095433122692, + "loss": 0.0614, + "num_input_tokens_seen": 20633824, + "step": 15910 + }, + { + "epoch": 0.7776121955390516, + "grad_norm": 0.0017324495129287243, + "learning_rate": 0.19731506035295046, + "loss": 0.12, + "num_input_tokens_seen": 20640512, + "step": 15915 + }, + { + "epoch": 0.777856497202746, + "grad_norm": 0.0008170511573553085, + "learning_rate": 0.19725915907809702, + "loss": 0.0946, + "num_input_tokens_seen": 20647040, + "step": 15920 + }, + { + "epoch": 0.7781007988664402, + "grad_norm": 0.0009171414421871305, + "learning_rate": 0.1972032505152874, + "loss": 0.0921, + "num_input_tokens_seen": 20653888, + "step": 15925 + }, + { + "epoch": 0.7783451005301346, + "grad_norm": 0.0013771748635917902, + "learning_rate": 0.19714733467314338, + "loss": 0.1046, + "num_input_tokens_seen": 20660448, + "step": 15930 + }, + { + "epoch": 0.778589402193829, + "grad_norm": 0.0011880755191668868, + "learning_rate": 0.19709141156028784, + "loss": 0.1178, + "num_input_tokens_seen": 20666208, + "step": 15935 + }, + { + "epoch": 0.7788337038575233, + "grad_norm": 0.0009610185516066849, + "learning_rate": 0.1970354811853448, + "loss": 0.0973, + "num_input_tokens_seen": 20672352, + "step": 15940 + }, + { + "epoch": 0.7790780055212176, + "grad_norm": 0.001892867498099804, + "learning_rate": 0.19697954355693953, + "loss": 0.092, + "num_input_tokens_seen": 20678592, + "step": 15945 + }, + { + "epoch": 0.7793223071849119, + "grad_norm": 0.0018828778993338346, + "learning_rate": 0.19692359868369827, + "loss": 0.0795, + "num_input_tokens_seen": 20685952, + "step": 15950 + }, + { + "epoch": 0.7795666088486063, + "grad_norm": 0.0012766438303515315, + "learning_rate": 0.1968676465742484, + "loss": 0.1131, + "num_input_tokens_seen": 20692224, + "step": 15955 + }, + { + "epoch": 0.7798109105123006, + "grad_norm": 0.001244527637027204, + "learning_rate": 0.19681168723721845, + "loss": 0.1314, + "num_input_tokens_seen": 20698592, + "step": 15960 + }, + { + "epoch": 0.7800552121759949, + "grad_norm": 0.0009122220217250288, + "learning_rate": 0.19675572068123803, + "loss": 0.0832, + "num_input_tokens_seen": 20705344, + "step": 15965 + }, + { + "epoch": 0.7802995138396892, + "grad_norm": 0.001260383054614067, + "learning_rate": 0.19669974691493794, + "loss": 0.1095, + "num_input_tokens_seen": 20711936, + "step": 15970 + }, + { + "epoch": 0.7805438155033836, + "grad_norm": 0.0010031993733718991, + "learning_rate": 0.19664376594695002, + "loss": 0.0986, + "num_input_tokens_seen": 20718112, + "step": 15975 + }, + { + "epoch": 0.7807881171670779, + "grad_norm": 0.001664525712840259, + "learning_rate": 0.19658777778590722, + "loss": 0.0972, + "num_input_tokens_seen": 20724288, + "step": 15980 + }, + { + "epoch": 0.7810324188307722, + "grad_norm": 0.0009739671368151903, + "learning_rate": 0.19653178244044364, + "loss": 0.0896, + "num_input_tokens_seen": 20730592, + "step": 15985 + }, + { + "epoch": 0.7812767204944666, + "grad_norm": 0.0017115657683461905, + "learning_rate": 0.19647577991919443, + "loss": 0.0989, + "num_input_tokens_seen": 20737280, + "step": 15990 + }, + { + "epoch": 0.7815210221581609, + "grad_norm": 0.0012450385838747025, + "learning_rate": 0.1964197702307959, + "loss": 0.0942, + "num_input_tokens_seen": 20743584, + "step": 15995 + }, + { + "epoch": 0.7817653238218553, + "grad_norm": 0.000984130660071969, + "learning_rate": 0.19636375338388545, + "loss": 0.0977, + "num_input_tokens_seen": 20750368, + "step": 16000 + }, + { + "epoch": 0.7817653238218553, + "eval_loss": 0.1014992967247963, + "eval_runtime": 402.1299, + "eval_samples_per_second": 90.481, + "eval_steps_per_second": 22.622, + "num_input_tokens_seen": 20750368, + "step": 16000 + }, + { + "epoch": 0.7820096254855495, + "grad_norm": 0.0008685843786224723, + "learning_rate": 0.1963077293871016, + "loss": 0.0882, + "num_input_tokens_seen": 20756736, + "step": 16005 + }, + { + "epoch": 0.7822539271492439, + "grad_norm": 0.001573816523887217, + "learning_rate": 0.19625169824908395, + "loss": 0.1211, + "num_input_tokens_seen": 20763136, + "step": 16010 + }, + { + "epoch": 0.7824982288129382, + "grad_norm": 0.0012744978303089738, + "learning_rate": 0.19619565997847319, + "loss": 0.112, + "num_input_tokens_seen": 20769344, + "step": 16015 + }, + { + "epoch": 0.7827425304766326, + "grad_norm": 0.001386958989314735, + "learning_rate": 0.19613961458391113, + "loss": 0.1076, + "num_input_tokens_seen": 20776256, + "step": 16020 + }, + { + "epoch": 0.7829868321403268, + "grad_norm": 0.0025690619368106127, + "learning_rate": 0.19608356207404065, + "loss": 0.1284, + "num_input_tokens_seen": 20782592, + "step": 16025 + }, + { + "epoch": 0.7832311338040212, + "grad_norm": 0.0010675664525479078, + "learning_rate": 0.1960275024575058, + "loss": 0.0699, + "num_input_tokens_seen": 20789376, + "step": 16030 + }, + { + "epoch": 0.7834754354677156, + "grad_norm": 0.0008892891346476972, + "learning_rate": 0.19597143574295164, + "loss": 0.1039, + "num_input_tokens_seen": 20795840, + "step": 16035 + }, + { + "epoch": 0.7837197371314099, + "grad_norm": 0.0013409120729193091, + "learning_rate": 0.1959153619390244, + "loss": 0.1249, + "num_input_tokens_seen": 20802624, + "step": 16040 + }, + { + "epoch": 0.7839640387951042, + "grad_norm": 0.0012580405455082655, + "learning_rate": 0.1958592810543713, + "loss": 0.1034, + "num_input_tokens_seen": 20808992, + "step": 16045 + }, + { + "epoch": 0.7842083404587985, + "grad_norm": 0.002273187506943941, + "learning_rate": 0.19580319309764077, + "loss": 0.1173, + "num_input_tokens_seen": 20815456, + "step": 16050 + }, + { + "epoch": 0.7844526421224929, + "grad_norm": 0.0015690414002165198, + "learning_rate": 0.1957470980774823, + "loss": 0.1196, + "num_input_tokens_seen": 20822432, + "step": 16055 + }, + { + "epoch": 0.7846969437861872, + "grad_norm": 0.0010692378273233771, + "learning_rate": 0.19569099600254639, + "loss": 0.1115, + "num_input_tokens_seen": 20828608, + "step": 16060 + }, + { + "epoch": 0.7849412454498815, + "grad_norm": 0.0015949906082823873, + "learning_rate": 0.1956348868814847, + "loss": 0.0942, + "num_input_tokens_seen": 20835360, + "step": 16065 + }, + { + "epoch": 0.7851855471135758, + "grad_norm": 0.0008672370458953083, + "learning_rate": 0.19557877072295, + "loss": 0.1077, + "num_input_tokens_seen": 20842144, + "step": 16070 + }, + { + "epoch": 0.7854298487772702, + "grad_norm": 0.0012733940966427326, + "learning_rate": 0.19552264753559603, + "loss": 0.1236, + "num_input_tokens_seen": 20848032, + "step": 16075 + }, + { + "epoch": 0.7856741504409646, + "grad_norm": 0.000870087300427258, + "learning_rate": 0.19546651732807774, + "loss": 0.091, + "num_input_tokens_seen": 20854400, + "step": 16080 + }, + { + "epoch": 0.7859184521046588, + "grad_norm": 0.0008117982069961727, + "learning_rate": 0.19541038010905112, + "loss": 0.0892, + "num_input_tokens_seen": 20861088, + "step": 16085 + }, + { + "epoch": 0.7861627537683532, + "grad_norm": 0.0009439666173420846, + "learning_rate": 0.19535423588717324, + "loss": 0.0918, + "num_input_tokens_seen": 20867168, + "step": 16090 + }, + { + "epoch": 0.7864070554320475, + "grad_norm": 0.0015956107527017593, + "learning_rate": 0.19529808467110224, + "loss": 0.1029, + "num_input_tokens_seen": 20874208, + "step": 16095 + }, + { + "epoch": 0.7866513570957419, + "grad_norm": 0.0010947075206786394, + "learning_rate": 0.19524192646949734, + "loss": 0.1216, + "num_input_tokens_seen": 20881088, + "step": 16100 + }, + { + "epoch": 0.7868956587594361, + "grad_norm": 0.0017259666929021478, + "learning_rate": 0.19518576129101878, + "loss": 0.1041, + "num_input_tokens_seen": 20887424, + "step": 16105 + }, + { + "epoch": 0.7871399604231305, + "grad_norm": 0.0013259021798148751, + "learning_rate": 0.19512958914432804, + "loss": 0.1072, + "num_input_tokens_seen": 20893792, + "step": 16110 + }, + { + "epoch": 0.7873842620868248, + "grad_norm": 0.0010426136432215571, + "learning_rate": 0.1950734100380875, + "loss": 0.1083, + "num_input_tokens_seen": 20899872, + "step": 16115 + }, + { + "epoch": 0.7876285637505192, + "grad_norm": 0.0016536528710275888, + "learning_rate": 0.19501722398096066, + "loss": 0.0763, + "num_input_tokens_seen": 20906208, + "step": 16120 + }, + { + "epoch": 0.7878728654142134, + "grad_norm": 0.0011509553296491504, + "learning_rate": 0.1949610309816122, + "loss": 0.1156, + "num_input_tokens_seen": 20912416, + "step": 16125 + }, + { + "epoch": 0.7881171670779078, + "grad_norm": 0.001299541792832315, + "learning_rate": 0.1949048310487078, + "loss": 0.0816, + "num_input_tokens_seen": 20919136, + "step": 16130 + }, + { + "epoch": 0.7883614687416022, + "grad_norm": 0.001573065179400146, + "learning_rate": 0.19484862419091406, + "loss": 0.1256, + "num_input_tokens_seen": 20925568, + "step": 16135 + }, + { + "epoch": 0.7886057704052964, + "grad_norm": 0.001341172493994236, + "learning_rate": 0.19479241041689893, + "loss": 0.068, + "num_input_tokens_seen": 20932064, + "step": 16140 + }, + { + "epoch": 0.7888500720689908, + "grad_norm": 0.0011315756710246205, + "learning_rate": 0.19473618973533116, + "loss": 0.1128, + "num_input_tokens_seen": 20938592, + "step": 16145 + }, + { + "epoch": 0.7890943737326851, + "grad_norm": 0.0015304816188290715, + "learning_rate": 0.19467996215488076, + "loss": 0.1307, + "num_input_tokens_seen": 20944832, + "step": 16150 + }, + { + "epoch": 0.7893386753963795, + "grad_norm": 0.0017190856160596013, + "learning_rate": 0.1946237276842187, + "loss": 0.1097, + "num_input_tokens_seen": 20951584, + "step": 16155 + }, + { + "epoch": 0.7895829770600737, + "grad_norm": 0.0006895111291669309, + "learning_rate": 0.19456748633201712, + "loss": 0.0932, + "num_input_tokens_seen": 20958336, + "step": 16160 + }, + { + "epoch": 0.7898272787237681, + "grad_norm": 0.0006020612781867385, + "learning_rate": 0.194511238106949, + "loss": 0.1131, + "num_input_tokens_seen": 20964512, + "step": 16165 + }, + { + "epoch": 0.7900715803874624, + "grad_norm": 0.0008329474949277937, + "learning_rate": 0.19445498301768863, + "loss": 0.0972, + "num_input_tokens_seen": 20970912, + "step": 16170 + }, + { + "epoch": 0.7903158820511568, + "grad_norm": 0.0008598589920438826, + "learning_rate": 0.19439872107291126, + "loss": 0.0798, + "num_input_tokens_seen": 20977408, + "step": 16175 + }, + { + "epoch": 0.7905601837148512, + "grad_norm": 0.0011450131423771381, + "learning_rate": 0.1943424522812931, + "loss": 0.0893, + "num_input_tokens_seen": 20983648, + "step": 16180 + }, + { + "epoch": 0.7908044853785454, + "grad_norm": 0.0009514742996543646, + "learning_rate": 0.19428617665151157, + "loss": 0.0888, + "num_input_tokens_seen": 20990432, + "step": 16185 + }, + { + "epoch": 0.7910487870422398, + "grad_norm": 0.0012400596169754863, + "learning_rate": 0.19422989419224507, + "loss": 0.1045, + "num_input_tokens_seen": 20997152, + "step": 16190 + }, + { + "epoch": 0.7912930887059341, + "grad_norm": 0.0010557003552094102, + "learning_rate": 0.19417360491217303, + "loss": 0.0971, + "num_input_tokens_seen": 21003776, + "step": 16195 + }, + { + "epoch": 0.7915373903696284, + "grad_norm": 0.0009881883161142468, + "learning_rate": 0.19411730881997605, + "loss": 0.1011, + "num_input_tokens_seen": 21010432, + "step": 16200 + }, + { + "epoch": 0.7915373903696284, + "eval_loss": 0.10021702200174332, + "eval_runtime": 401.8638, + "eval_samples_per_second": 90.541, + "eval_steps_per_second": 22.637, + "num_input_tokens_seen": 21010432, + "step": 16200 + }, + { + "epoch": 0.7917816920333227, + "grad_norm": 0.0013421521289274096, + "learning_rate": 0.1940610059243356, + "loss": 0.0937, + "num_input_tokens_seen": 21016736, + "step": 16205 + }, + { + "epoch": 0.7920259936970171, + "grad_norm": 0.001448864466510713, + "learning_rate": 0.19400469623393435, + "loss": 0.1155, + "num_input_tokens_seen": 21023776, + "step": 16210 + }, + { + "epoch": 0.7922702953607114, + "grad_norm": 0.001400207867845893, + "learning_rate": 0.1939483797574559, + "loss": 0.1076, + "num_input_tokens_seen": 21030016, + "step": 16215 + }, + { + "epoch": 0.7925145970244057, + "grad_norm": 0.0008748463587835431, + "learning_rate": 0.19389205650358504, + "loss": 0.0629, + "num_input_tokens_seen": 21036832, + "step": 16220 + }, + { + "epoch": 0.7927588986881, + "grad_norm": 0.0022219656966626644, + "learning_rate": 0.19383572648100747, + "loss": 0.0942, + "num_input_tokens_seen": 21043264, + "step": 16225 + }, + { + "epoch": 0.7930032003517944, + "grad_norm": 0.001831138739362359, + "learning_rate": 0.19377938969841, + "loss": 0.0902, + "num_input_tokens_seen": 21049824, + "step": 16230 + }, + { + "epoch": 0.7932475020154888, + "grad_norm": 0.0015721910167485476, + "learning_rate": 0.1937230461644805, + "loss": 0.1173, + "num_input_tokens_seen": 21056640, + "step": 16235 + }, + { + "epoch": 0.793491803679183, + "grad_norm": 0.0007913807057775557, + "learning_rate": 0.19366669588790777, + "loss": 0.1092, + "num_input_tokens_seen": 21063360, + "step": 16240 + }, + { + "epoch": 0.7937361053428774, + "grad_norm": 0.0011833420721814036, + "learning_rate": 0.19361033887738185, + "loss": 0.1134, + "num_input_tokens_seen": 21069824, + "step": 16245 + }, + { + "epoch": 0.7939804070065717, + "grad_norm": 0.0008583606104366481, + "learning_rate": 0.19355397514159361, + "loss": 0.0693, + "num_input_tokens_seen": 21076416, + "step": 16250 + }, + { + "epoch": 0.7942247086702661, + "grad_norm": 0.0011405842378735542, + "learning_rate": 0.19349760468923508, + "loss": 0.074, + "num_input_tokens_seen": 21082496, + "step": 16255 + }, + { + "epoch": 0.7944690103339603, + "grad_norm": 0.0016460232436656952, + "learning_rate": 0.19344122752899925, + "loss": 0.1075, + "num_input_tokens_seen": 21088992, + "step": 16260 + }, + { + "epoch": 0.7947133119976547, + "grad_norm": 0.0009800223633646965, + "learning_rate": 0.1933848436695802, + "loss": 0.0825, + "num_input_tokens_seen": 21095328, + "step": 16265 + }, + { + "epoch": 0.794957613661349, + "grad_norm": 0.0015283257234841585, + "learning_rate": 0.1933284531196731, + "loss": 0.0899, + "num_input_tokens_seen": 21101440, + "step": 16270 + }, + { + "epoch": 0.7952019153250434, + "grad_norm": 0.0024028352927416563, + "learning_rate": 0.19327205588797403, + "loss": 0.108, + "num_input_tokens_seen": 21108064, + "step": 16275 + }, + { + "epoch": 0.7954462169887377, + "grad_norm": 0.0016526625258848071, + "learning_rate": 0.19321565198318014, + "loss": 0.1343, + "num_input_tokens_seen": 21114464, + "step": 16280 + }, + { + "epoch": 0.795690518652432, + "grad_norm": 0.001132941571995616, + "learning_rate": 0.1931592414139896, + "loss": 0.1007, + "num_input_tokens_seen": 21121568, + "step": 16285 + }, + { + "epoch": 0.7959348203161264, + "grad_norm": 0.0017407401464879513, + "learning_rate": 0.19310282418910169, + "loss": 0.0866, + "num_input_tokens_seen": 21127968, + "step": 16290 + }, + { + "epoch": 0.7961791219798207, + "grad_norm": 0.001348821329884231, + "learning_rate": 0.1930464003172166, + "loss": 0.0945, + "num_input_tokens_seen": 21133920, + "step": 16295 + }, + { + "epoch": 0.796423423643515, + "grad_norm": 0.002263967413455248, + "learning_rate": 0.19298996980703567, + "loss": 0.1101, + "num_input_tokens_seen": 21140672, + "step": 16300 + }, + { + "epoch": 0.7966677253072093, + "grad_norm": 0.0011036793002858758, + "learning_rate": 0.19293353266726113, + "loss": 0.099, + "num_input_tokens_seen": 21147168, + "step": 16305 + }, + { + "epoch": 0.7969120269709037, + "grad_norm": 0.0007948071579448879, + "learning_rate": 0.19287708890659633, + "loss": 0.102, + "num_input_tokens_seen": 21153664, + "step": 16310 + }, + { + "epoch": 0.797156328634598, + "grad_norm": 0.001300059724599123, + "learning_rate": 0.19282063853374556, + "loss": 0.0903, + "num_input_tokens_seen": 21160320, + "step": 16315 + }, + { + "epoch": 0.7974006302982923, + "grad_norm": 0.0007893629954196513, + "learning_rate": 0.19276418155741423, + "loss": 0.0751, + "num_input_tokens_seen": 21166816, + "step": 16320 + }, + { + "epoch": 0.7976449319619867, + "grad_norm": 0.0008373786695301533, + "learning_rate": 0.19270771798630867, + "loss": 0.1024, + "num_input_tokens_seen": 21173504, + "step": 16325 + }, + { + "epoch": 0.797889233625681, + "grad_norm": 0.001108943484723568, + "learning_rate": 0.1926512478291363, + "loss": 0.0571, + "num_input_tokens_seen": 21179936, + "step": 16330 + }, + { + "epoch": 0.7981335352893754, + "grad_norm": 0.001695183222182095, + "learning_rate": 0.19259477109460557, + "loss": 0.0918, + "num_input_tokens_seen": 21186368, + "step": 16335 + }, + { + "epoch": 0.7983778369530696, + "grad_norm": 0.0014088003663346171, + "learning_rate": 0.19253828779142584, + "loss": 0.0932, + "num_input_tokens_seen": 21192704, + "step": 16340 + }, + { + "epoch": 0.798622138616764, + "grad_norm": 0.0013085320824757218, + "learning_rate": 0.19248179792830755, + "loss": 0.0919, + "num_input_tokens_seen": 21199200, + "step": 16345 + }, + { + "epoch": 0.7988664402804583, + "grad_norm": 0.0021353510674089193, + "learning_rate": 0.19242530151396217, + "loss": 0.1329, + "num_input_tokens_seen": 21205088, + "step": 16350 + }, + { + "epoch": 0.7991107419441527, + "grad_norm": 0.0018231645226478577, + "learning_rate": 0.19236879855710215, + "loss": 0.1215, + "num_input_tokens_seen": 21211040, + "step": 16355 + }, + { + "epoch": 0.7993550436078469, + "grad_norm": 0.001580537180416286, + "learning_rate": 0.19231228906644096, + "loss": 0.1178, + "num_input_tokens_seen": 21217216, + "step": 16360 + }, + { + "epoch": 0.7995993452715413, + "grad_norm": 0.0010945259127765894, + "learning_rate": 0.19225577305069302, + "loss": 0.1038, + "num_input_tokens_seen": 21223936, + "step": 16365 + }, + { + "epoch": 0.7998436469352356, + "grad_norm": 0.0020145620219409466, + "learning_rate": 0.1921992505185739, + "loss": 0.0909, + "num_input_tokens_seen": 21230496, + "step": 16370 + }, + { + "epoch": 0.80008794859893, + "grad_norm": 0.0009404998854734004, + "learning_rate": 0.19214272147880004, + "loss": 0.107, + "num_input_tokens_seen": 21236672, + "step": 16375 + }, + { + "epoch": 0.8003322502626243, + "grad_norm": 0.0013725970638915896, + "learning_rate": 0.19208618594008892, + "loss": 0.0829, + "num_input_tokens_seen": 21243744, + "step": 16380 + }, + { + "epoch": 0.8005765519263186, + "grad_norm": 0.0010799859883263707, + "learning_rate": 0.19202964391115904, + "loss": 0.1026, + "num_input_tokens_seen": 21250304, + "step": 16385 + }, + { + "epoch": 0.800820853590013, + "grad_norm": 0.0015132800908759236, + "learning_rate": 0.1919730954007299, + "loss": 0.1298, + "num_input_tokens_seen": 21256864, + "step": 16390 + }, + { + "epoch": 0.8010651552537073, + "grad_norm": 0.0013646666193380952, + "learning_rate": 0.19191654041752199, + "loss": 0.1092, + "num_input_tokens_seen": 21263456, + "step": 16395 + }, + { + "epoch": 0.8013094569174016, + "grad_norm": 0.0010950065916404128, + "learning_rate": 0.19185997897025678, + "loss": 0.0978, + "num_input_tokens_seen": 21270112, + "step": 16400 + }, + { + "epoch": 0.8013094569174016, + "eval_loss": 0.10168339312076569, + "eval_runtime": 402.2793, + "eval_samples_per_second": 90.447, + "eval_steps_per_second": 22.614, + "num_input_tokens_seen": 21270112, + "step": 16400 + }, + { + "epoch": 0.8015537585810959, + "grad_norm": 0.0008846784476190805, + "learning_rate": 0.19180341106765672, + "loss": 0.09, + "num_input_tokens_seen": 21276864, + "step": 16405 + }, + { + "epoch": 0.8017980602447903, + "grad_norm": 0.0008386471890844405, + "learning_rate": 0.19174683671844536, + "loss": 0.0834, + "num_input_tokens_seen": 21283392, + "step": 16410 + }, + { + "epoch": 0.8020423619084845, + "grad_norm": 0.0017781833885237575, + "learning_rate": 0.19169025593134717, + "loss": 0.0833, + "num_input_tokens_seen": 21290048, + "step": 16415 + }, + { + "epoch": 0.8022866635721789, + "grad_norm": 0.0023815545719116926, + "learning_rate": 0.19163366871508764, + "loss": 0.1226, + "num_input_tokens_seen": 21296704, + "step": 16420 + }, + { + "epoch": 0.8025309652358733, + "grad_norm": 0.0008876225911080837, + "learning_rate": 0.19157707507839317, + "loss": 0.0861, + "num_input_tokens_seen": 21303456, + "step": 16425 + }, + { + "epoch": 0.8027752668995676, + "grad_norm": 0.0014530763728544116, + "learning_rate": 0.19152047502999123, + "loss": 0.102, + "num_input_tokens_seen": 21309792, + "step": 16430 + }, + { + "epoch": 0.803019568563262, + "grad_norm": 0.0014155457029119134, + "learning_rate": 0.19146386857861025, + "loss": 0.0881, + "num_input_tokens_seen": 21316544, + "step": 16435 + }, + { + "epoch": 0.8032638702269562, + "grad_norm": 0.0017629819922149181, + "learning_rate": 0.19140725573297968, + "loss": 0.1276, + "num_input_tokens_seen": 21322816, + "step": 16440 + }, + { + "epoch": 0.8035081718906506, + "grad_norm": 0.0013832347467541695, + "learning_rate": 0.19135063650182987, + "loss": 0.106, + "num_input_tokens_seen": 21329120, + "step": 16445 + }, + { + "epoch": 0.8037524735543449, + "grad_norm": 0.0009964711498469114, + "learning_rate": 0.19129401089389234, + "loss": 0.0974, + "num_input_tokens_seen": 21335904, + "step": 16450 + }, + { + "epoch": 0.8039967752180393, + "grad_norm": 0.0012162381317466497, + "learning_rate": 0.19123737891789938, + "loss": 0.1099, + "num_input_tokens_seen": 21342368, + "step": 16455 + }, + { + "epoch": 0.8042410768817335, + "grad_norm": 0.0008037079824134707, + "learning_rate": 0.19118074058258439, + "loss": 0.1075, + "num_input_tokens_seen": 21348896, + "step": 16460 + }, + { + "epoch": 0.8044853785454279, + "grad_norm": 0.0017932202899828553, + "learning_rate": 0.1911240958966816, + "loss": 0.1213, + "num_input_tokens_seen": 21355168, + "step": 16465 + }, + { + "epoch": 0.8047296802091223, + "grad_norm": 0.0010266980389133096, + "learning_rate": 0.19106744486892652, + "loss": 0.1105, + "num_input_tokens_seen": 21361952, + "step": 16470 + }, + { + "epoch": 0.8049739818728165, + "grad_norm": 0.0015837869141250849, + "learning_rate": 0.1910107875080553, + "loss": 0.1087, + "num_input_tokens_seen": 21368000, + "step": 16475 + }, + { + "epoch": 0.8052182835365109, + "grad_norm": 0.00130571355111897, + "learning_rate": 0.19095412382280533, + "loss": 0.0824, + "num_input_tokens_seen": 21374784, + "step": 16480 + }, + { + "epoch": 0.8054625852002052, + "grad_norm": 0.0008377988706342876, + "learning_rate": 0.19089745382191473, + "loss": 0.1259, + "num_input_tokens_seen": 21381568, + "step": 16485 + }, + { + "epoch": 0.8057068868638996, + "grad_norm": 0.0009452843223698437, + "learning_rate": 0.19084077751412284, + "loss": 0.0947, + "num_input_tokens_seen": 21387904, + "step": 16490 + }, + { + "epoch": 0.8059511885275938, + "grad_norm": 0.0006980644538998604, + "learning_rate": 0.19078409490816986, + "loss": 0.0795, + "num_input_tokens_seen": 21394208, + "step": 16495 + }, + { + "epoch": 0.8061954901912882, + "grad_norm": 0.002207131590694189, + "learning_rate": 0.19072740601279686, + "loss": 0.0961, + "num_input_tokens_seen": 21400576, + "step": 16500 + }, + { + "epoch": 0.8064397918549825, + "grad_norm": 0.0008252050611190498, + "learning_rate": 0.19067071083674605, + "loss": 0.0873, + "num_input_tokens_seen": 21406944, + "step": 16505 + }, + { + "epoch": 0.8066840935186769, + "grad_norm": 0.001216157921589911, + "learning_rate": 0.19061400938876052, + "loss": 0.0758, + "num_input_tokens_seen": 21413408, + "step": 16510 + }, + { + "epoch": 0.8069283951823711, + "grad_norm": 0.0010278256377205253, + "learning_rate": 0.1905573016775844, + "loss": 0.0798, + "num_input_tokens_seen": 21419488, + "step": 16515 + }, + { + "epoch": 0.8071726968460655, + "grad_norm": 0.0024372818879783154, + "learning_rate": 0.19050058771196263, + "loss": 0.1101, + "num_input_tokens_seen": 21425984, + "step": 16520 + }, + { + "epoch": 0.8074169985097599, + "grad_norm": 0.0012670040596276522, + "learning_rate": 0.19044386750064132, + "loss": 0.1163, + "num_input_tokens_seen": 21432544, + "step": 16525 + }, + { + "epoch": 0.8076613001734542, + "grad_norm": 0.0015893530799075961, + "learning_rate": 0.19038714105236737, + "loss": 0.1072, + "num_input_tokens_seen": 21439200, + "step": 16530 + }, + { + "epoch": 0.8079056018371485, + "grad_norm": 0.0015380795812234282, + "learning_rate": 0.19033040837588874, + "loss": 0.107, + "num_input_tokens_seen": 21446272, + "step": 16535 + }, + { + "epoch": 0.8081499035008428, + "grad_norm": 0.0011865677079185843, + "learning_rate": 0.1902736694799543, + "loss": 0.0816, + "num_input_tokens_seen": 21452800, + "step": 16540 + }, + { + "epoch": 0.8083942051645372, + "grad_norm": 0.0014394884929060936, + "learning_rate": 0.19021692437331392, + "loss": 0.104, + "num_input_tokens_seen": 21459328, + "step": 16545 + }, + { + "epoch": 0.8086385068282315, + "grad_norm": 0.0010696286335587502, + "learning_rate": 0.1901601730647184, + "loss": 0.0987, + "num_input_tokens_seen": 21465408, + "step": 16550 + }, + { + "epoch": 0.8088828084919258, + "grad_norm": 0.0013010856928303838, + "learning_rate": 0.19010341556291954, + "loss": 0.0948, + "num_input_tokens_seen": 21472224, + "step": 16555 + }, + { + "epoch": 0.8091271101556201, + "grad_norm": 0.0011434074258431792, + "learning_rate": 0.19004665187667, + "loss": 0.1069, + "num_input_tokens_seen": 21479456, + "step": 16560 + }, + { + "epoch": 0.8093714118193145, + "grad_norm": 0.0010049162665382028, + "learning_rate": 0.1899898820147235, + "loss": 0.1136, + "num_input_tokens_seen": 21485728, + "step": 16565 + }, + { + "epoch": 0.8096157134830089, + "grad_norm": 0.0010016578016802669, + "learning_rate": 0.18993310598583465, + "loss": 0.0897, + "num_input_tokens_seen": 21492320, + "step": 16570 + }, + { + "epoch": 0.8098600151467031, + "grad_norm": 0.0016396790742874146, + "learning_rate": 0.18987632379875904, + "loss": 0.1183, + "num_input_tokens_seen": 21498720, + "step": 16575 + }, + { + "epoch": 0.8101043168103975, + "grad_norm": 0.0008923118002712727, + "learning_rate": 0.18981953546225314, + "loss": 0.0972, + "num_input_tokens_seen": 21505568, + "step": 16580 + }, + { + "epoch": 0.8103486184740918, + "grad_norm": 0.0005828440771438181, + "learning_rate": 0.18976274098507445, + "loss": 0.1128, + "num_input_tokens_seen": 21512096, + "step": 16585 + }, + { + "epoch": 0.8105929201377862, + "grad_norm": 0.0007879590848460793, + "learning_rate": 0.18970594037598146, + "loss": 0.0702, + "num_input_tokens_seen": 21518784, + "step": 16590 + }, + { + "epoch": 0.8108372218014804, + "grad_norm": 0.002055126940831542, + "learning_rate": 0.1896491336437335, + "loss": 0.0973, + "num_input_tokens_seen": 21525024, + "step": 16595 + }, + { + "epoch": 0.8110815234651748, + "grad_norm": 0.0016066604293882847, + "learning_rate": 0.18959232079709085, + "loss": 0.1306, + "num_input_tokens_seen": 21531456, + "step": 16600 + }, + { + "epoch": 0.8110815234651748, + "eval_loss": 0.10309136658906937, + "eval_runtime": 402.1895, + "eval_samples_per_second": 90.467, + "eval_steps_per_second": 22.619, + "num_input_tokens_seen": 21531456, + "step": 16600 + }, + { + "epoch": 0.8113258251288691, + "grad_norm": 0.0011388794519007206, + "learning_rate": 0.18953550184481477, + "loss": 0.0884, + "num_input_tokens_seen": 21537888, + "step": 16605 + }, + { + "epoch": 0.8115701267925635, + "grad_norm": 0.0013549481518566608, + "learning_rate": 0.18947867679566752, + "loss": 0.1088, + "num_input_tokens_seen": 21544288, + "step": 16610 + }, + { + "epoch": 0.8118144284562578, + "grad_norm": 0.001494531985372305, + "learning_rate": 0.18942184565841216, + "loss": 0.1052, + "num_input_tokens_seen": 21550784, + "step": 16615 + }, + { + "epoch": 0.8120587301199521, + "grad_norm": 0.0010344479233026505, + "learning_rate": 0.18936500844181278, + "loss": 0.0976, + "num_input_tokens_seen": 21557280, + "step": 16620 + }, + { + "epoch": 0.8123030317836465, + "grad_norm": 0.0011006041895598173, + "learning_rate": 0.18930816515463436, + "loss": 0.1045, + "num_input_tokens_seen": 21563808, + "step": 16625 + }, + { + "epoch": 0.8125473334473408, + "grad_norm": 0.001731606898829341, + "learning_rate": 0.18925131580564297, + "loss": 0.1063, + "num_input_tokens_seen": 21570656, + "step": 16630 + }, + { + "epoch": 0.8127916351110351, + "grad_norm": 0.0012589580146595836, + "learning_rate": 0.1891944604036054, + "loss": 0.1121, + "num_input_tokens_seen": 21576736, + "step": 16635 + }, + { + "epoch": 0.8130359367747294, + "grad_norm": 0.000921903585549444, + "learning_rate": 0.1891375989572895, + "loss": 0.1028, + "num_input_tokens_seen": 21583424, + "step": 16640 + }, + { + "epoch": 0.8132802384384238, + "grad_norm": 0.0009147649398073554, + "learning_rate": 0.18908073147546398, + "loss": 0.0752, + "num_input_tokens_seen": 21590048, + "step": 16645 + }, + { + "epoch": 0.813524540102118, + "grad_norm": 0.000999247538857162, + "learning_rate": 0.18902385796689858, + "loss": 0.0755, + "num_input_tokens_seen": 21596288, + "step": 16650 + }, + { + "epoch": 0.8137688417658124, + "grad_norm": 0.0013791468227282166, + "learning_rate": 0.18896697844036384, + "loss": 0.107, + "num_input_tokens_seen": 21603008, + "step": 16655 + }, + { + "epoch": 0.8140131434295067, + "grad_norm": 0.0015919465804472566, + "learning_rate": 0.18891009290463137, + "loss": 0.1043, + "num_input_tokens_seen": 21609248, + "step": 16660 + }, + { + "epoch": 0.8142574450932011, + "grad_norm": 0.0011910966131836176, + "learning_rate": 0.18885320136847353, + "loss": 0.1315, + "num_input_tokens_seen": 21615424, + "step": 16665 + }, + { + "epoch": 0.8145017467568955, + "grad_norm": 0.0009967322694137692, + "learning_rate": 0.1887963038406639, + "loss": 0.0968, + "num_input_tokens_seen": 21621760, + "step": 16670 + }, + { + "epoch": 0.8147460484205897, + "grad_norm": 0.0016979691572487354, + "learning_rate": 0.18873940032997658, + "loss": 0.1104, + "num_input_tokens_seen": 21627872, + "step": 16675 + }, + { + "epoch": 0.8149903500842841, + "grad_norm": 0.0007171880570240319, + "learning_rate": 0.18868249084518693, + "loss": 0.1079, + "num_input_tokens_seen": 21634560, + "step": 16680 + }, + { + "epoch": 0.8152346517479784, + "grad_norm": 0.001158380415290594, + "learning_rate": 0.18862557539507102, + "loss": 0.0938, + "num_input_tokens_seen": 21640960, + "step": 16685 + }, + { + "epoch": 0.8154789534116728, + "grad_norm": 0.0012445999309420586, + "learning_rate": 0.18856865398840605, + "loss": 0.1381, + "num_input_tokens_seen": 21646976, + "step": 16690 + }, + { + "epoch": 0.815723255075367, + "grad_norm": 0.001033117761835456, + "learning_rate": 0.18851172663396995, + "loss": 0.0768, + "num_input_tokens_seen": 21653440, + "step": 16695 + }, + { + "epoch": 0.8159675567390614, + "grad_norm": 0.0021194766741245985, + "learning_rate": 0.1884547933405416, + "loss": 0.1166, + "num_input_tokens_seen": 21659776, + "step": 16700 + }, + { + "epoch": 0.8162118584027557, + "grad_norm": 0.0007428288226947188, + "learning_rate": 0.1883978541169009, + "loss": 0.087, + "num_input_tokens_seen": 21666272, + "step": 16705 + }, + { + "epoch": 0.81645616006645, + "grad_norm": 0.001122209825553, + "learning_rate": 0.18834090897182854, + "loss": 0.0803, + "num_input_tokens_seen": 21672640, + "step": 16710 + }, + { + "epoch": 0.8167004617301444, + "grad_norm": 0.0008840240770950913, + "learning_rate": 0.1882839579141062, + "loss": 0.108, + "num_input_tokens_seen": 21678848, + "step": 16715 + }, + { + "epoch": 0.8169447633938387, + "grad_norm": 0.001513768802396953, + "learning_rate": 0.18822700095251646, + "loss": 0.1123, + "num_input_tokens_seen": 21685152, + "step": 16720 + }, + { + "epoch": 0.8171890650575331, + "grad_norm": 0.0015589771792292595, + "learning_rate": 0.18817003809584273, + "loss": 0.0927, + "num_input_tokens_seen": 21692064, + "step": 16725 + }, + { + "epoch": 0.8174333667212274, + "grad_norm": 0.0010501215001568198, + "learning_rate": 0.1881130693528695, + "loss": 0.0945, + "num_input_tokens_seen": 21698432, + "step": 16730 + }, + { + "epoch": 0.8176776683849217, + "grad_norm": 0.0009075297857634723, + "learning_rate": 0.18805609473238197, + "loss": 0.106, + "num_input_tokens_seen": 21705088, + "step": 16735 + }, + { + "epoch": 0.817921970048616, + "grad_norm": 0.0009181052446365356, + "learning_rate": 0.18799911424316643, + "loss": 0.1017, + "num_input_tokens_seen": 21711712, + "step": 16740 + }, + { + "epoch": 0.8181662717123104, + "grad_norm": 0.0009607138927094638, + "learning_rate": 0.18794212789400994, + "loss": 0.1011, + "num_input_tokens_seen": 21718432, + "step": 16745 + }, + { + "epoch": 0.8184105733760046, + "grad_norm": 0.0017339674523100257, + "learning_rate": 0.18788513569370052, + "loss": 0.1022, + "num_input_tokens_seen": 21724800, + "step": 16750 + }, + { + "epoch": 0.818654875039699, + "grad_norm": 0.000926917651668191, + "learning_rate": 0.1878281376510271, + "loss": 0.1097, + "num_input_tokens_seen": 21731264, + "step": 16755 + }, + { + "epoch": 0.8188991767033934, + "grad_norm": 0.001195400720462203, + "learning_rate": 0.18777113377477941, + "loss": 0.095, + "num_input_tokens_seen": 21737568, + "step": 16760 + }, + { + "epoch": 0.8191434783670877, + "grad_norm": 0.0010377186117693782, + "learning_rate": 0.1877141240737483, + "loss": 0.093, + "num_input_tokens_seen": 21743936, + "step": 16765 + }, + { + "epoch": 0.819387780030782, + "grad_norm": 0.0012795700458809733, + "learning_rate": 0.18765710855672527, + "loss": 0.095, + "num_input_tokens_seen": 21751040, + "step": 16770 + }, + { + "epoch": 0.8196320816944763, + "grad_norm": 0.0014428958529606462, + "learning_rate": 0.18760008723250288, + "loss": 0.0992, + "num_input_tokens_seen": 21757312, + "step": 16775 + }, + { + "epoch": 0.8198763833581707, + "grad_norm": 0.0007904673111625016, + "learning_rate": 0.18754306010987457, + "loss": 0.0944, + "num_input_tokens_seen": 21763424, + "step": 16780 + }, + { + "epoch": 0.820120685021865, + "grad_norm": 0.0020543814171105623, + "learning_rate": 0.18748602719763457, + "loss": 0.1086, + "num_input_tokens_seen": 21769504, + "step": 16785 + }, + { + "epoch": 0.8203649866855593, + "grad_norm": 0.0012502263998612761, + "learning_rate": 0.18742898850457804, + "loss": 0.1135, + "num_input_tokens_seen": 21775936, + "step": 16790 + }, + { + "epoch": 0.8206092883492536, + "grad_norm": 0.0007043506484478712, + "learning_rate": 0.1873719440395012, + "loss": 0.0939, + "num_input_tokens_seen": 21782560, + "step": 16795 + }, + { + "epoch": 0.820853590012948, + "grad_norm": 0.000517923675943166, + "learning_rate": 0.1873148938112009, + "loss": 0.0884, + "num_input_tokens_seen": 21788384, + "step": 16800 + }, + { + "epoch": 0.820853590012948, + "eval_loss": 0.10469167679548264, + "eval_runtime": 402.632, + "eval_samples_per_second": 90.368, + "eval_steps_per_second": 22.594, + "num_input_tokens_seen": 21788384, + "step": 16800 + }, + { + "epoch": 0.8210978916766423, + "grad_norm": 0.0009373110951855779, + "learning_rate": 0.18725783782847508, + "loss": 0.1251, + "num_input_tokens_seen": 21794944, + "step": 16805 + }, + { + "epoch": 0.8213421933403366, + "grad_norm": 0.0007497326005250216, + "learning_rate": 0.1872007761001224, + "loss": 0.0772, + "num_input_tokens_seen": 21801664, + "step": 16810 + }, + { + "epoch": 0.821586495004031, + "grad_norm": 0.001502069178968668, + "learning_rate": 0.1871437086349426, + "loss": 0.1142, + "num_input_tokens_seen": 21807872, + "step": 16815 + }, + { + "epoch": 0.8218307966677253, + "grad_norm": 0.0012109157396480441, + "learning_rate": 0.18708663544173615, + "loss": 0.1025, + "num_input_tokens_seen": 21814432, + "step": 16820 + }, + { + "epoch": 0.8220750983314197, + "grad_norm": 0.000964490172918886, + "learning_rate": 0.18702955652930442, + "loss": 0.1023, + "num_input_tokens_seen": 21820896, + "step": 16825 + }, + { + "epoch": 0.8223193999951139, + "grad_norm": 0.0011809486895799637, + "learning_rate": 0.18697247190644972, + "loss": 0.0849, + "num_input_tokens_seen": 21827168, + "step": 16830 + }, + { + "epoch": 0.8225637016588083, + "grad_norm": 0.0013292429503053427, + "learning_rate": 0.18691538158197527, + "loss": 0.1, + "num_input_tokens_seen": 21833728, + "step": 16835 + }, + { + "epoch": 0.8228080033225026, + "grad_norm": 0.001564305741339922, + "learning_rate": 0.1868582855646851, + "loss": 0.1138, + "num_input_tokens_seen": 21839968, + "step": 16840 + }, + { + "epoch": 0.823052304986197, + "grad_norm": 0.0009694506879895926, + "learning_rate": 0.18680118386338404, + "loss": 0.0902, + "num_input_tokens_seen": 21846432, + "step": 16845 + }, + { + "epoch": 0.8232966066498912, + "grad_norm": 0.0010340071748942137, + "learning_rate": 0.18674407648687794, + "loss": 0.1041, + "num_input_tokens_seen": 21852608, + "step": 16850 + }, + { + "epoch": 0.8235409083135856, + "grad_norm": 0.00123350671492517, + "learning_rate": 0.1866869634439736, + "loss": 0.1322, + "num_input_tokens_seen": 21858912, + "step": 16855 + }, + { + "epoch": 0.82378520997728, + "grad_norm": 0.0013617750955745578, + "learning_rate": 0.18662984474347838, + "loss": 0.0806, + "num_input_tokens_seen": 21865440, + "step": 16860 + }, + { + "epoch": 0.8240295116409743, + "grad_norm": 0.0014885287964716554, + "learning_rate": 0.1865727203942008, + "loss": 0.0915, + "num_input_tokens_seen": 21871776, + "step": 16865 + }, + { + "epoch": 0.8242738133046686, + "grad_norm": 0.0012869597412645817, + "learning_rate": 0.1865155904049501, + "loss": 0.0902, + "num_input_tokens_seen": 21877856, + "step": 16870 + }, + { + "epoch": 0.8245181149683629, + "grad_norm": 0.0007502001826651394, + "learning_rate": 0.1864584547845365, + "loss": 0.0728, + "num_input_tokens_seen": 21884192, + "step": 16875 + }, + { + "epoch": 0.8247624166320573, + "grad_norm": 0.0010703742736950517, + "learning_rate": 0.186401313541771, + "loss": 0.111, + "num_input_tokens_seen": 21890944, + "step": 16880 + }, + { + "epoch": 0.8250067182957516, + "grad_norm": 0.0016990953590720892, + "learning_rate": 0.18634416668546552, + "loss": 0.1304, + "num_input_tokens_seen": 21897536, + "step": 16885 + }, + { + "epoch": 0.8252510199594459, + "grad_norm": 0.0010923153022304177, + "learning_rate": 0.1862870142244328, + "loss": 0.0948, + "num_input_tokens_seen": 21904192, + "step": 16890 + }, + { + "epoch": 0.8254953216231402, + "grad_norm": 0.0014199558645486832, + "learning_rate": 0.1862298561674865, + "loss": 0.1143, + "num_input_tokens_seen": 21910560, + "step": 16895 + }, + { + "epoch": 0.8257396232868346, + "grad_norm": 0.001702694920822978, + "learning_rate": 0.18617269252344104, + "loss": 0.1283, + "num_input_tokens_seen": 21916928, + "step": 16900 + }, + { + "epoch": 0.8259839249505289, + "grad_norm": 0.0007763081812299788, + "learning_rate": 0.18611552330111186, + "loss": 0.0991, + "num_input_tokens_seen": 21923264, + "step": 16905 + }, + { + "epoch": 0.8262282266142232, + "grad_norm": 0.0015705700498074293, + "learning_rate": 0.18605834850931507, + "loss": 0.0929, + "num_input_tokens_seen": 21929888, + "step": 16910 + }, + { + "epoch": 0.8264725282779176, + "grad_norm": 0.001093815895728767, + "learning_rate": 0.18600116815686787, + "loss": 0.0815, + "num_input_tokens_seen": 21936576, + "step": 16915 + }, + { + "epoch": 0.8267168299416119, + "grad_norm": 0.0007145747076719999, + "learning_rate": 0.1859439822525881, + "loss": 0.0921, + "num_input_tokens_seen": 21942848, + "step": 16920 + }, + { + "epoch": 0.8269611316053063, + "grad_norm": 0.0012240047799423337, + "learning_rate": 0.18588679080529455, + "loss": 0.0952, + "num_input_tokens_seen": 21949184, + "step": 16925 + }, + { + "epoch": 0.8272054332690005, + "grad_norm": 0.00109349365811795, + "learning_rate": 0.1858295938238069, + "loss": 0.0912, + "num_input_tokens_seen": 21956096, + "step": 16930 + }, + { + "epoch": 0.8274497349326949, + "grad_norm": 0.0011158449342474341, + "learning_rate": 0.18577239131694562, + "loss": 0.1322, + "num_input_tokens_seen": 21962752, + "step": 16935 + }, + { + "epoch": 0.8276940365963892, + "grad_norm": 0.0013526866678148508, + "learning_rate": 0.18571518329353204, + "loss": 0.108, + "num_input_tokens_seen": 21968832, + "step": 16940 + }, + { + "epoch": 0.8279383382600836, + "grad_norm": 0.0012275681365281343, + "learning_rate": 0.18565796976238838, + "loss": 0.0711, + "num_input_tokens_seen": 21974944, + "step": 16945 + }, + { + "epoch": 0.8281826399237778, + "grad_norm": 0.0008314584847539663, + "learning_rate": 0.18560075073233764, + "loss": 0.1081, + "num_input_tokens_seen": 21981568, + "step": 16950 + }, + { + "epoch": 0.8284269415874722, + "grad_norm": 0.0008963873842731118, + "learning_rate": 0.18554352621220377, + "loss": 0.0665, + "num_input_tokens_seen": 21988640, + "step": 16955 + }, + { + "epoch": 0.8286712432511666, + "grad_norm": 0.0010320756118744612, + "learning_rate": 0.18548629621081153, + "loss": 0.0876, + "num_input_tokens_seen": 21995264, + "step": 16960 + }, + { + "epoch": 0.8289155449148609, + "grad_norm": 0.0016006234800443053, + "learning_rate": 0.18542906073698645, + "loss": 0.1089, + "num_input_tokens_seen": 22001248, + "step": 16965 + }, + { + "epoch": 0.8291598465785552, + "grad_norm": 0.0009842050494626164, + "learning_rate": 0.18537181979955494, + "loss": 0.1065, + "num_input_tokens_seen": 22007328, + "step": 16970 + }, + { + "epoch": 0.8294041482422495, + "grad_norm": 0.001006507663987577, + "learning_rate": 0.18531457340734434, + "loss": 0.0913, + "num_input_tokens_seen": 22013760, + "step": 16975 + }, + { + "epoch": 0.8296484499059439, + "grad_norm": 0.0014101857086643577, + "learning_rate": 0.1852573215691827, + "loss": 0.0675, + "num_input_tokens_seen": 22020256, + "step": 16980 + }, + { + "epoch": 0.8298927515696382, + "grad_norm": 0.0014858419308438897, + "learning_rate": 0.18520006429389904, + "loss": 0.0935, + "num_input_tokens_seen": 22026624, + "step": 16985 + }, + { + "epoch": 0.8301370532333325, + "grad_norm": 0.0013130364241078496, + "learning_rate": 0.1851428015903231, + "loss": 0.1264, + "num_input_tokens_seen": 22032800, + "step": 16990 + }, + { + "epoch": 0.8303813548970268, + "grad_norm": 0.0010369418887421489, + "learning_rate": 0.1850855334672855, + "loss": 0.0879, + "num_input_tokens_seen": 22039264, + "step": 16995 + }, + { + "epoch": 0.8306256565607212, + "grad_norm": 0.0011605763575062156, + "learning_rate": 0.1850282599336178, + "loss": 0.0938, + "num_input_tokens_seen": 22045600, + "step": 17000 + }, + { + "epoch": 0.8306256565607212, + "eval_loss": 0.10154983401298523, + "eval_runtime": 402.731, + "eval_samples_per_second": 90.346, + "eval_steps_per_second": 22.588, + "num_input_tokens_seen": 22045600, + "step": 17000 + }, + { + "epoch": 0.8308699582244156, + "grad_norm": 0.001355719636194408, + "learning_rate": 0.18497098099815215, + "loss": 0.1252, + "num_input_tokens_seen": 22051904, + "step": 17005 + }, + { + "epoch": 0.8311142598881098, + "grad_norm": 0.0016480369959026575, + "learning_rate": 0.18491369666972174, + "loss": 0.1103, + "num_input_tokens_seen": 22058528, + "step": 17010 + }, + { + "epoch": 0.8313585615518042, + "grad_norm": 0.001086199888959527, + "learning_rate": 0.1848564069571606, + "loss": 0.1041, + "num_input_tokens_seen": 22065408, + "step": 17015 + }, + { + "epoch": 0.8316028632154985, + "grad_norm": 0.0009097747388295829, + "learning_rate": 0.18479911186930348, + "loss": 0.1043, + "num_input_tokens_seen": 22071680, + "step": 17020 + }, + { + "epoch": 0.8318471648791929, + "grad_norm": 0.0018615551525726914, + "learning_rate": 0.18474181141498597, + "loss": 0.1005, + "num_input_tokens_seen": 22078144, + "step": 17025 + }, + { + "epoch": 0.8320914665428871, + "grad_norm": 0.0012526497012004256, + "learning_rate": 0.18468450560304453, + "loss": 0.0989, + "num_input_tokens_seen": 22084512, + "step": 17030 + }, + { + "epoch": 0.8323357682065815, + "grad_norm": 0.0009159085457213223, + "learning_rate": 0.1846271944423165, + "loss": 0.1138, + "num_input_tokens_seen": 22090816, + "step": 17035 + }, + { + "epoch": 0.8325800698702758, + "grad_norm": 0.0008768619736656547, + "learning_rate": 0.18456987794163993, + "loss": 0.0865, + "num_input_tokens_seen": 22097248, + "step": 17040 + }, + { + "epoch": 0.8328243715339702, + "grad_norm": 0.0009028617641888559, + "learning_rate": 0.18451255610985373, + "loss": 0.1156, + "num_input_tokens_seen": 22103360, + "step": 17045 + }, + { + "epoch": 0.8330686731976644, + "grad_norm": 0.0012691965093836188, + "learning_rate": 0.18445522895579766, + "loss": 0.1181, + "num_input_tokens_seen": 22109568, + "step": 17050 + }, + { + "epoch": 0.8333129748613588, + "grad_norm": 0.000674959272146225, + "learning_rate": 0.1843978964883123, + "loss": 0.0844, + "num_input_tokens_seen": 22115968, + "step": 17055 + }, + { + "epoch": 0.8335572765250532, + "grad_norm": 0.0011496428633108735, + "learning_rate": 0.18434055871623906, + "loss": 0.0837, + "num_input_tokens_seen": 22122240, + "step": 17060 + }, + { + "epoch": 0.8338015781887474, + "grad_norm": 0.0019191191531717777, + "learning_rate": 0.18428321564842007, + "loss": 0.1131, + "num_input_tokens_seen": 22129280, + "step": 17065 + }, + { + "epoch": 0.8340458798524418, + "grad_norm": 0.0011215067934244871, + "learning_rate": 0.18422586729369841, + "loss": 0.1059, + "num_input_tokens_seen": 22135264, + "step": 17070 + }, + { + "epoch": 0.8342901815161361, + "grad_norm": 0.0015695752808824182, + "learning_rate": 0.1841685136609179, + "loss": 0.1084, + "num_input_tokens_seen": 22141824, + "step": 17075 + }, + { + "epoch": 0.8345344831798305, + "grad_norm": 0.000875564175657928, + "learning_rate": 0.18411115475892326, + "loss": 0.0962, + "num_input_tokens_seen": 22147840, + "step": 17080 + }, + { + "epoch": 0.8347787848435247, + "grad_norm": 0.0011298086028546095, + "learning_rate": 0.18405379059655982, + "loss": 0.0991, + "num_input_tokens_seen": 22154496, + "step": 17085 + }, + { + "epoch": 0.8350230865072191, + "grad_norm": 0.0007289340137504041, + "learning_rate": 0.1839964211826739, + "loss": 0.0883, + "num_input_tokens_seen": 22161184, + "step": 17090 + }, + { + "epoch": 0.8352673881709134, + "grad_norm": 0.000942873302847147, + "learning_rate": 0.18393904652611265, + "loss": 0.0895, + "num_input_tokens_seen": 22167840, + "step": 17095 + }, + { + "epoch": 0.8355116898346078, + "grad_norm": 0.0008676146971993148, + "learning_rate": 0.18388166663572392, + "loss": 0.0972, + "num_input_tokens_seen": 22173824, + "step": 17100 + }, + { + "epoch": 0.8357559914983022, + "grad_norm": 0.0012650835560634732, + "learning_rate": 0.18382428152035643, + "loss": 0.0994, + "num_input_tokens_seen": 22180576, + "step": 17105 + }, + { + "epoch": 0.8360002931619964, + "grad_norm": 0.0013618589146062732, + "learning_rate": 0.1837668911888596, + "loss": 0.0928, + "num_input_tokens_seen": 22187424, + "step": 17110 + }, + { + "epoch": 0.8362445948256908, + "grad_norm": 0.0018439808627590537, + "learning_rate": 0.18370949565008388, + "loss": 0.1216, + "num_input_tokens_seen": 22193568, + "step": 17115 + }, + { + "epoch": 0.8364888964893851, + "grad_norm": 0.0010952359298244119, + "learning_rate": 0.1836520949128803, + "loss": 0.1033, + "num_input_tokens_seen": 22199968, + "step": 17120 + }, + { + "epoch": 0.8367331981530794, + "grad_norm": 0.0014886059798300266, + "learning_rate": 0.18359468898610076, + "loss": 0.1045, + "num_input_tokens_seen": 22206496, + "step": 17125 + }, + { + "epoch": 0.8369774998167737, + "grad_norm": 0.0018095406703650951, + "learning_rate": 0.18353727787859797, + "loss": 0.126, + "num_input_tokens_seen": 22212928, + "step": 17130 + }, + { + "epoch": 0.8372218014804681, + "grad_norm": 0.0013352453242987394, + "learning_rate": 0.18347986159922552, + "loss": 0.0927, + "num_input_tokens_seen": 22220032, + "step": 17135 + }, + { + "epoch": 0.8374661031441624, + "grad_norm": 0.0011292354902252555, + "learning_rate": 0.1834224401568377, + "loss": 0.1018, + "num_input_tokens_seen": 22226016, + "step": 17140 + }, + { + "epoch": 0.8377104048078567, + "grad_norm": 0.0009545779903419316, + "learning_rate": 0.1833650135602896, + "loss": 0.1347, + "num_input_tokens_seen": 22232224, + "step": 17145 + }, + { + "epoch": 0.8379547064715511, + "grad_norm": 0.0020334243308752775, + "learning_rate": 0.18330758181843707, + "loss": 0.1256, + "num_input_tokens_seen": 22238464, + "step": 17150 + }, + { + "epoch": 0.8381990081352454, + "grad_norm": 0.0011899527162313461, + "learning_rate": 0.18325014494013686, + "loss": 0.1043, + "num_input_tokens_seen": 22244512, + "step": 17155 + }, + { + "epoch": 0.8384433097989398, + "grad_norm": 0.0012869302881881595, + "learning_rate": 0.18319270293424647, + "loss": 0.0905, + "num_input_tokens_seen": 22251328, + "step": 17160 + }, + { + "epoch": 0.838687611462634, + "grad_norm": 0.0012457637349143624, + "learning_rate": 0.18313525580962417, + "loss": 0.0916, + "num_input_tokens_seen": 22257888, + "step": 17165 + }, + { + "epoch": 0.8389319131263284, + "grad_norm": 0.0013576822821050882, + "learning_rate": 0.18307780357512896, + "loss": 0.0948, + "num_input_tokens_seen": 22264544, + "step": 17170 + }, + { + "epoch": 0.8391762147900227, + "grad_norm": 0.0015651665162295103, + "learning_rate": 0.1830203462396208, + "loss": 0.1162, + "num_input_tokens_seen": 22271328, + "step": 17175 + }, + { + "epoch": 0.8394205164537171, + "grad_norm": 0.000798265275079757, + "learning_rate": 0.18296288381196033, + "loss": 0.0886, + "num_input_tokens_seen": 22278016, + "step": 17180 + }, + { + "epoch": 0.8396648181174113, + "grad_norm": 0.0016458200989291072, + "learning_rate": 0.1829054163010089, + "loss": 0.1, + "num_input_tokens_seen": 22284544, + "step": 17185 + }, + { + "epoch": 0.8399091197811057, + "grad_norm": 0.0011667523067444563, + "learning_rate": 0.18284794371562874, + "loss": 0.0841, + "num_input_tokens_seen": 22291072, + "step": 17190 + }, + { + "epoch": 0.8401534214448, + "grad_norm": 0.0011990853818133473, + "learning_rate": 0.18279046606468288, + "loss": 0.0909, + "num_input_tokens_seen": 22297216, + "step": 17195 + }, + { + "epoch": 0.8403977231084944, + "grad_norm": 0.0010203799465671182, + "learning_rate": 0.1827329833570351, + "loss": 0.089, + "num_input_tokens_seen": 22303808, + "step": 17200 + }, + { + "epoch": 0.8403977231084944, + "eval_loss": 0.10059717297554016, + "eval_runtime": 402.1412, + "eval_samples_per_second": 90.478, + "eval_steps_per_second": 22.621, + "num_input_tokens_seen": 22303808, + "step": 17200 + }, + { + "epoch": 0.8406420247721887, + "grad_norm": 0.0010918519692495465, + "learning_rate": 0.18267549560154991, + "loss": 0.1212, + "num_input_tokens_seen": 22310272, + "step": 17205 + }, + { + "epoch": 0.840886326435883, + "grad_norm": 0.0009949643863365054, + "learning_rate": 0.18261800280709267, + "loss": 0.1034, + "num_input_tokens_seen": 22316640, + "step": 17210 + }, + { + "epoch": 0.8411306280995774, + "grad_norm": 0.0009817165555432439, + "learning_rate": 0.18256050498252957, + "loss": 0.1109, + "num_input_tokens_seen": 22323872, + "step": 17215 + }, + { + "epoch": 0.8413749297632717, + "grad_norm": 0.0009013881208375096, + "learning_rate": 0.18250300213672735, + "loss": 0.1035, + "num_input_tokens_seen": 22330080, + "step": 17220 + }, + { + "epoch": 0.841619231426966, + "grad_norm": 0.0008223050972446799, + "learning_rate": 0.18244549427855378, + "loss": 0.1082, + "num_input_tokens_seen": 22336320, + "step": 17225 + }, + { + "epoch": 0.8418635330906603, + "grad_norm": 0.001343710464425385, + "learning_rate": 0.1823879814168772, + "loss": 0.0978, + "num_input_tokens_seen": 22342688, + "step": 17230 + }, + { + "epoch": 0.8421078347543547, + "grad_norm": 0.0008468664600513875, + "learning_rate": 0.18233046356056692, + "loss": 0.096, + "num_input_tokens_seen": 22348992, + "step": 17235 + }, + { + "epoch": 0.842352136418049, + "grad_norm": 0.0008437005453743041, + "learning_rate": 0.18227294071849284, + "loss": 0.102, + "num_input_tokens_seen": 22355552, + "step": 17240 + }, + { + "epoch": 0.8425964380817433, + "grad_norm": 0.0012129091192036867, + "learning_rate": 0.18221541289952578, + "loss": 0.0728, + "num_input_tokens_seen": 22362528, + "step": 17245 + }, + { + "epoch": 0.8428407397454377, + "grad_norm": 0.0022569752763956785, + "learning_rate": 0.18215788011253717, + "loss": 0.1489, + "num_input_tokens_seen": 22369280, + "step": 17250 + }, + { + "epoch": 0.843085041409132, + "grad_norm": 0.0011304912623018026, + "learning_rate": 0.18210034236639935, + "loss": 0.0786, + "num_input_tokens_seen": 22376096, + "step": 17255 + }, + { + "epoch": 0.8433293430728264, + "grad_norm": 0.001540794619359076, + "learning_rate": 0.1820427996699853, + "loss": 0.1039, + "num_input_tokens_seen": 22382528, + "step": 17260 + }, + { + "epoch": 0.8435736447365206, + "grad_norm": 0.0018824859289452434, + "learning_rate": 0.1819852520321689, + "loss": 0.1146, + "num_input_tokens_seen": 22388768, + "step": 17265 + }, + { + "epoch": 0.843817946400215, + "grad_norm": 0.0027522281743586063, + "learning_rate": 0.18192769946182466, + "loss": 0.0965, + "num_input_tokens_seen": 22395328, + "step": 17270 + }, + { + "epoch": 0.8440622480639093, + "grad_norm": 0.0012321349931880832, + "learning_rate": 0.18187014196782794, + "loss": 0.0791, + "num_input_tokens_seen": 22401792, + "step": 17275 + }, + { + "epoch": 0.8443065497276037, + "grad_norm": 0.0012179380282759666, + "learning_rate": 0.18181257955905486, + "loss": 0.0815, + "num_input_tokens_seen": 22408832, + "step": 17280 + }, + { + "epoch": 0.8445508513912979, + "grad_norm": 0.0008465931168757379, + "learning_rate": 0.18175501224438217, + "loss": 0.1044, + "num_input_tokens_seen": 22415264, + "step": 17285 + }, + { + "epoch": 0.8447951530549923, + "grad_norm": 0.0010317414999008179, + "learning_rate": 0.18169744003268756, + "loss": 0.1093, + "num_input_tokens_seen": 22421600, + "step": 17290 + }, + { + "epoch": 0.8450394547186867, + "grad_norm": 0.0012693445896729827, + "learning_rate": 0.18163986293284937, + "loss": 0.1016, + "num_input_tokens_seen": 22428096, + "step": 17295 + }, + { + "epoch": 0.845283756382381, + "grad_norm": 0.0023849254939705133, + "learning_rate": 0.18158228095374673, + "loss": 0.1072, + "num_input_tokens_seen": 22434560, + "step": 17300 + }, + { + "epoch": 0.8455280580460753, + "grad_norm": 0.000962122343480587, + "learning_rate": 0.18152469410425945, + "loss": 0.0816, + "num_input_tokens_seen": 22440576, + "step": 17305 + }, + { + "epoch": 0.8457723597097696, + "grad_norm": 0.0014891799073666334, + "learning_rate": 0.18146710239326813, + "loss": 0.0996, + "num_input_tokens_seen": 22447104, + "step": 17310 + }, + { + "epoch": 0.846016661373464, + "grad_norm": 0.0012012326624244452, + "learning_rate": 0.18140950582965423, + "loss": 0.0904, + "num_input_tokens_seen": 22453440, + "step": 17315 + }, + { + "epoch": 0.8462609630371583, + "grad_norm": 0.0008964978042058647, + "learning_rate": 0.1813519044222998, + "loss": 0.1013, + "num_input_tokens_seen": 22459584, + "step": 17320 + }, + { + "epoch": 0.8465052647008526, + "grad_norm": 0.0017495477804914117, + "learning_rate": 0.18129429818008772, + "loss": 0.1078, + "num_input_tokens_seen": 22466112, + "step": 17325 + }, + { + "epoch": 0.8467495663645469, + "grad_norm": 0.000792598002590239, + "learning_rate": 0.18123668711190163, + "loss": 0.108, + "num_input_tokens_seen": 22472768, + "step": 17330 + }, + { + "epoch": 0.8469938680282413, + "grad_norm": 0.0017144136363640428, + "learning_rate": 0.18117907122662583, + "loss": 0.0987, + "num_input_tokens_seen": 22479232, + "step": 17335 + }, + { + "epoch": 0.8472381696919355, + "grad_norm": 0.0018728113500401378, + "learning_rate": 0.1811214505331454, + "loss": 0.1098, + "num_input_tokens_seen": 22485888, + "step": 17340 + }, + { + "epoch": 0.8474824713556299, + "grad_norm": 0.0016248130705207586, + "learning_rate": 0.1810638250403462, + "loss": 0.0925, + "num_input_tokens_seen": 22491904, + "step": 17345 + }, + { + "epoch": 0.8477267730193243, + "grad_norm": 0.0010431921109557152, + "learning_rate": 0.1810061947571148, + "loss": 0.1009, + "num_input_tokens_seen": 22498016, + "step": 17350 + }, + { + "epoch": 0.8479710746830186, + "grad_norm": 0.000896542682312429, + "learning_rate": 0.1809485596923385, + "loss": 0.0673, + "num_input_tokens_seen": 22504192, + "step": 17355 + }, + { + "epoch": 0.848215376346713, + "grad_norm": 0.000598429178353399, + "learning_rate": 0.18089091985490546, + "loss": 0.1138, + "num_input_tokens_seen": 22510496, + "step": 17360 + }, + { + "epoch": 0.8484596780104072, + "grad_norm": 0.0014486615546047688, + "learning_rate": 0.18083327525370432, + "loss": 0.0925, + "num_input_tokens_seen": 22516800, + "step": 17365 + }, + { + "epoch": 0.8487039796741016, + "grad_norm": 0.0007049728883430362, + "learning_rate": 0.18077562589762464, + "loss": 0.0831, + "num_input_tokens_seen": 22523168, + "step": 17370 + }, + { + "epoch": 0.8489482813377959, + "grad_norm": 0.0011834875913336873, + "learning_rate": 0.1807179717955567, + "loss": 0.1128, + "num_input_tokens_seen": 22529856, + "step": 17375 + }, + { + "epoch": 0.8491925830014903, + "grad_norm": 0.0010023314971476793, + "learning_rate": 0.1806603129563915, + "loss": 0.1023, + "num_input_tokens_seen": 22536192, + "step": 17380 + }, + { + "epoch": 0.8494368846651845, + "grad_norm": 0.001318457187153399, + "learning_rate": 0.1806026493890208, + "loss": 0.0736, + "num_input_tokens_seen": 22542848, + "step": 17385 + }, + { + "epoch": 0.8496811863288789, + "grad_norm": 0.00129300681874156, + "learning_rate": 0.18054498110233688, + "loss": 0.1082, + "num_input_tokens_seen": 22549312, + "step": 17390 + }, + { + "epoch": 0.8499254879925733, + "grad_norm": 0.0011321650817990303, + "learning_rate": 0.1804873081052331, + "loss": 0.1174, + "num_input_tokens_seen": 22556032, + "step": 17395 + }, + { + "epoch": 0.8501697896562675, + "grad_norm": 0.0009337373194284737, + "learning_rate": 0.18042963040660326, + "loss": 0.0837, + "num_input_tokens_seen": 22562496, + "step": 17400 + }, + { + "epoch": 0.8501697896562675, + "eval_loss": 0.10001740604639053, + "eval_runtime": 402.649, + "eval_samples_per_second": 90.364, + "eval_steps_per_second": 22.593, + "num_input_tokens_seen": 22562496, + "step": 17400 + }, + { + "epoch": 0.8504140913199619, + "grad_norm": 0.001261001219972968, + "learning_rate": 0.180371948015342, + "loss": 0.1152, + "num_input_tokens_seen": 22568864, + "step": 17405 + }, + { + "epoch": 0.8506583929836562, + "grad_norm": 0.0009654082241468132, + "learning_rate": 0.18031426094034472, + "loss": 0.0936, + "num_input_tokens_seen": 22575328, + "step": 17410 + }, + { + "epoch": 0.8509026946473506, + "grad_norm": 0.0010676439851522446, + "learning_rate": 0.18025656919050737, + "loss": 0.0832, + "num_input_tokens_seen": 22582048, + "step": 17415 + }, + { + "epoch": 0.8511469963110448, + "grad_norm": 0.0011191746452823281, + "learning_rate": 0.18019887277472688, + "loss": 0.1033, + "num_input_tokens_seen": 22588256, + "step": 17420 + }, + { + "epoch": 0.8513912979747392, + "grad_norm": 0.002137997653335333, + "learning_rate": 0.18014117170190067, + "loss": 0.1265, + "num_input_tokens_seen": 22594880, + "step": 17425 + }, + { + "epoch": 0.8516355996384335, + "grad_norm": 0.0012763352133333683, + "learning_rate": 0.18008346598092703, + "loss": 0.0765, + "num_input_tokens_seen": 22601376, + "step": 17430 + }, + { + "epoch": 0.8518799013021279, + "grad_norm": 0.0010515436297282577, + "learning_rate": 0.18002575562070489, + "loss": 0.0973, + "num_input_tokens_seen": 22608032, + "step": 17435 + }, + { + "epoch": 0.8521242029658221, + "grad_norm": 0.0011224944610148668, + "learning_rate": 0.1799680406301339, + "loss": 0.0991, + "num_input_tokens_seen": 22614752, + "step": 17440 + }, + { + "epoch": 0.8523685046295165, + "grad_norm": 0.0015787755837664008, + "learning_rate": 0.17991032101811447, + "loss": 0.0962, + "num_input_tokens_seen": 22621056, + "step": 17445 + }, + { + "epoch": 0.8526128062932109, + "grad_norm": 0.0018486428307369351, + "learning_rate": 0.1798525967935476, + "loss": 0.1328, + "num_input_tokens_seen": 22627584, + "step": 17450 + }, + { + "epoch": 0.8528571079569052, + "grad_norm": 0.001061272923834622, + "learning_rate": 0.17979486796533517, + "loss": 0.092, + "num_input_tokens_seen": 22633888, + "step": 17455 + }, + { + "epoch": 0.8531014096205995, + "grad_norm": 0.0011801565997302532, + "learning_rate": 0.1797371345423797, + "loss": 0.0843, + "num_input_tokens_seen": 22640480, + "step": 17460 + }, + { + "epoch": 0.8533457112842938, + "grad_norm": 0.0007609779131598771, + "learning_rate": 0.17967939653358436, + "loss": 0.0893, + "num_input_tokens_seen": 22647200, + "step": 17465 + }, + { + "epoch": 0.8535900129479882, + "grad_norm": 0.0008197049610316753, + "learning_rate": 0.17962165394785315, + "loss": 0.0945, + "num_input_tokens_seen": 22653280, + "step": 17470 + }, + { + "epoch": 0.8538343146116825, + "grad_norm": 0.0010261666029691696, + "learning_rate": 0.17956390679409057, + "loss": 0.112, + "num_input_tokens_seen": 22659552, + "step": 17475 + }, + { + "epoch": 0.8540786162753768, + "grad_norm": 0.0013354889815673232, + "learning_rate": 0.1795061550812021, + "loss": 0.1082, + "num_input_tokens_seen": 22666688, + "step": 17480 + }, + { + "epoch": 0.8543229179390711, + "grad_norm": 0.0011447608703747392, + "learning_rate": 0.1794483988180937, + "loss": 0.1188, + "num_input_tokens_seen": 22672960, + "step": 17485 + }, + { + "epoch": 0.8545672196027655, + "grad_norm": 0.0015280620427802205, + "learning_rate": 0.17939063801367214, + "loss": 0.1098, + "num_input_tokens_seen": 22679264, + "step": 17490 + }, + { + "epoch": 0.8548115212664599, + "grad_norm": 0.0009322873665951192, + "learning_rate": 0.17933287267684483, + "loss": 0.0824, + "num_input_tokens_seen": 22686528, + "step": 17495 + }, + { + "epoch": 0.8550558229301541, + "grad_norm": 0.0008819569484330714, + "learning_rate": 0.17927510281651995, + "loss": 0.0905, + "num_input_tokens_seen": 22692640, + "step": 17500 + }, + { + "epoch": 0.8553001245938485, + "grad_norm": 0.0006247460842132568, + "learning_rate": 0.17921732844160634, + "loss": 0.0882, + "num_input_tokens_seen": 22699392, + "step": 17505 + }, + { + "epoch": 0.8555444262575428, + "grad_norm": 0.0009700792143121362, + "learning_rate": 0.17915954956101351, + "loss": 0.1262, + "num_input_tokens_seen": 22705824, + "step": 17510 + }, + { + "epoch": 0.8557887279212372, + "grad_norm": 0.0011568529298529029, + "learning_rate": 0.17910176618365165, + "loss": 0.0837, + "num_input_tokens_seen": 22712320, + "step": 17515 + }, + { + "epoch": 0.8560330295849314, + "grad_norm": 0.001209837500937283, + "learning_rate": 0.17904397831843177, + "loss": 0.0991, + "num_input_tokens_seen": 22718720, + "step": 17520 + }, + { + "epoch": 0.8562773312486258, + "grad_norm": 0.0017048829467967153, + "learning_rate": 0.17898618597426547, + "loss": 0.1085, + "num_input_tokens_seen": 22725024, + "step": 17525 + }, + { + "epoch": 0.8565216329123201, + "grad_norm": 0.0016811022069305182, + "learning_rate": 0.17892838916006495, + "loss": 0.0882, + "num_input_tokens_seen": 22731488, + "step": 17530 + }, + { + "epoch": 0.8567659345760145, + "grad_norm": 0.0011322357458993793, + "learning_rate": 0.17887058788474333, + "loss": 0.0914, + "num_input_tokens_seen": 22737952, + "step": 17535 + }, + { + "epoch": 0.8570102362397088, + "grad_norm": 0.0013453017454594374, + "learning_rate": 0.17881278215721427, + "loss": 0.0885, + "num_input_tokens_seen": 22744256, + "step": 17540 + }, + { + "epoch": 0.8572545379034031, + "grad_norm": 0.0019411672838032246, + "learning_rate": 0.1787549719863921, + "loss": 0.0969, + "num_input_tokens_seen": 22750688, + "step": 17545 + }, + { + "epoch": 0.8574988395670975, + "grad_norm": 0.0010220528347417712, + "learning_rate": 0.17869715738119188, + "loss": 0.0998, + "num_input_tokens_seen": 22756704, + "step": 17550 + }, + { + "epoch": 0.8577431412307918, + "grad_norm": 0.0007568321889266372, + "learning_rate": 0.17863933835052936, + "loss": 0.0971, + "num_input_tokens_seen": 22762848, + "step": 17555 + }, + { + "epoch": 0.8579874428944861, + "grad_norm": 0.0005837763310410082, + "learning_rate": 0.17858151490332097, + "loss": 0.0732, + "num_input_tokens_seen": 22769216, + "step": 17560 + }, + { + "epoch": 0.8582317445581804, + "grad_norm": 0.0015011039795354009, + "learning_rate": 0.17852368704848381, + "loss": 0.1012, + "num_input_tokens_seen": 22775648, + "step": 17565 + }, + { + "epoch": 0.8584760462218748, + "grad_norm": 0.0008893508347682655, + "learning_rate": 0.17846585479493565, + "loss": 0.1271, + "num_input_tokens_seen": 22782016, + "step": 17570 + }, + { + "epoch": 0.8587203478855691, + "grad_norm": 0.0012076256098225713, + "learning_rate": 0.178408018151595, + "loss": 0.0817, + "num_input_tokens_seen": 22788448, + "step": 17575 + }, + { + "epoch": 0.8589646495492634, + "grad_norm": 0.0019219410605728626, + "learning_rate": 0.17835017712738085, + "loss": 0.1104, + "num_input_tokens_seen": 22794816, + "step": 17580 + }, + { + "epoch": 0.8592089512129577, + "grad_norm": 0.0009238149505108595, + "learning_rate": 0.17829233173121323, + "loss": 0.0981, + "num_input_tokens_seen": 22801536, + "step": 17585 + }, + { + "epoch": 0.8594532528766521, + "grad_norm": 0.0008604919421486557, + "learning_rate": 0.17823448197201244, + "loss": 0.0995, + "num_input_tokens_seen": 22808224, + "step": 17590 + }, + { + "epoch": 0.8596975545403465, + "grad_norm": 0.0008287224918603897, + "learning_rate": 0.1781766278586997, + "loss": 0.1008, + "num_input_tokens_seen": 22814400, + "step": 17595 + }, + { + "epoch": 0.8599418562040407, + "grad_norm": 0.0020857478957623243, + "learning_rate": 0.1781187694001969, + "loss": 0.0792, + "num_input_tokens_seen": 22821376, + "step": 17600 + }, + { + "epoch": 0.8599418562040407, + "eval_loss": 0.1037045568227768, + "eval_runtime": 402.7753, + "eval_samples_per_second": 90.336, + "eval_steps_per_second": 22.586, + "num_input_tokens_seen": 22821376, + "step": 17600 + }, + { + "epoch": 0.8601861578677351, + "grad_norm": 0.0009975965367630124, + "learning_rate": 0.1780609066054265, + "loss": 0.0653, + "num_input_tokens_seen": 22827840, + "step": 17605 + }, + { + "epoch": 0.8604304595314294, + "grad_norm": 0.0015268933493644, + "learning_rate": 0.17800303948331164, + "loss": 0.1125, + "num_input_tokens_seen": 22834144, + "step": 17610 + }, + { + "epoch": 0.8606747611951238, + "grad_norm": 0.0015686426777392626, + "learning_rate": 0.1779451680427762, + "loss": 0.1025, + "num_input_tokens_seen": 22840864, + "step": 17615 + }, + { + "epoch": 0.860919062858818, + "grad_norm": 0.0011061789700761437, + "learning_rate": 0.17788729229274464, + "loss": 0.0952, + "num_input_tokens_seen": 22847360, + "step": 17620 + }, + { + "epoch": 0.8611633645225124, + "grad_norm": 0.0007040568743832409, + "learning_rate": 0.17782941224214222, + "loss": 0.0716, + "num_input_tokens_seen": 22854048, + "step": 17625 + }, + { + "epoch": 0.8614076661862067, + "grad_norm": 0.0011908886954188347, + "learning_rate": 0.17777152789989464, + "loss": 0.0934, + "num_input_tokens_seen": 22860576, + "step": 17630 + }, + { + "epoch": 0.8616519678499011, + "grad_norm": 0.0029957403894513845, + "learning_rate": 0.17771363927492845, + "loss": 0.1133, + "num_input_tokens_seen": 22867168, + "step": 17635 + }, + { + "epoch": 0.8618962695135954, + "grad_norm": 0.0013865355867892504, + "learning_rate": 0.17765574637617085, + "loss": 0.0999, + "num_input_tokens_seen": 22873312, + "step": 17640 + }, + { + "epoch": 0.8621405711772897, + "grad_norm": 0.001481848070397973, + "learning_rate": 0.17759784921254962, + "loss": 0.1262, + "num_input_tokens_seen": 22879840, + "step": 17645 + }, + { + "epoch": 0.8623848728409841, + "grad_norm": 0.0009298375225625932, + "learning_rate": 0.1775399477929932, + "loss": 0.084, + "num_input_tokens_seen": 22886528, + "step": 17650 + }, + { + "epoch": 0.8626291745046784, + "grad_norm": 0.0007625161670148373, + "learning_rate": 0.17748204212643076, + "loss": 0.1182, + "num_input_tokens_seen": 22892704, + "step": 17655 + }, + { + "epoch": 0.8628734761683727, + "grad_norm": 0.0005585961043834686, + "learning_rate": 0.17742413222179204, + "loss": 0.0978, + "num_input_tokens_seen": 22899104, + "step": 17660 + }, + { + "epoch": 0.863117777832067, + "grad_norm": 0.00197678548283875, + "learning_rate": 0.17736621808800754, + "loss": 0.1083, + "num_input_tokens_seen": 22905728, + "step": 17665 + }, + { + "epoch": 0.8633620794957614, + "grad_norm": 0.0007598400115966797, + "learning_rate": 0.17730829973400827, + "loss": 0.09, + "num_input_tokens_seen": 22912096, + "step": 17670 + }, + { + "epoch": 0.8636063811594556, + "grad_norm": 0.0020493073388934135, + "learning_rate": 0.17725037716872602, + "loss": 0.1318, + "num_input_tokens_seen": 22918336, + "step": 17675 + }, + { + "epoch": 0.86385068282315, + "grad_norm": 0.001643909141421318, + "learning_rate": 0.17719245040109313, + "loss": 0.1184, + "num_input_tokens_seen": 22925024, + "step": 17680 + }, + { + "epoch": 0.8640949844868444, + "grad_norm": 0.0008611989323981106, + "learning_rate": 0.17713451944004271, + "loss": 0.0829, + "num_input_tokens_seen": 22931584, + "step": 17685 + }, + { + "epoch": 0.8643392861505387, + "grad_norm": 0.0010439656907692552, + "learning_rate": 0.17707658429450843, + "loss": 0.0846, + "num_input_tokens_seen": 22938368, + "step": 17690 + }, + { + "epoch": 0.8645835878142331, + "grad_norm": 0.0015968783991411328, + "learning_rate": 0.1770186449734245, + "loss": 0.0868, + "num_input_tokens_seen": 22945088, + "step": 17695 + }, + { + "epoch": 0.8648278894779273, + "grad_norm": 0.0010624118149280548, + "learning_rate": 0.17696070148572599, + "loss": 0.0912, + "num_input_tokens_seen": 22951488, + "step": 17700 + }, + { + "epoch": 0.8650721911416217, + "grad_norm": 0.002043432556092739, + "learning_rate": 0.17690275384034856, + "loss": 0.1269, + "num_input_tokens_seen": 22957888, + "step": 17705 + }, + { + "epoch": 0.865316492805316, + "grad_norm": 0.0014874530024826527, + "learning_rate": 0.17684480204622835, + "loss": 0.107, + "num_input_tokens_seen": 22964544, + "step": 17710 + }, + { + "epoch": 0.8655607944690104, + "grad_norm": 0.001311705564148724, + "learning_rate": 0.1767868461123023, + "loss": 0.097, + "num_input_tokens_seen": 22971328, + "step": 17715 + }, + { + "epoch": 0.8658050961327046, + "grad_norm": 0.0011263159103691578, + "learning_rate": 0.176728886047508, + "loss": 0.099, + "num_input_tokens_seen": 22977792, + "step": 17720 + }, + { + "epoch": 0.866049397796399, + "grad_norm": 0.0012308862060308456, + "learning_rate": 0.17667092186078362, + "loss": 0.1108, + "num_input_tokens_seen": 22984128, + "step": 17725 + }, + { + "epoch": 0.8662936994600933, + "grad_norm": 0.001366287819109857, + "learning_rate": 0.17661295356106785, + "loss": 0.1128, + "num_input_tokens_seen": 22990304, + "step": 17730 + }, + { + "epoch": 0.8665380011237876, + "grad_norm": 0.0010604987619444728, + "learning_rate": 0.1765549811573002, + "loss": 0.0929, + "num_input_tokens_seen": 22996800, + "step": 17735 + }, + { + "epoch": 0.866782302787482, + "grad_norm": 0.00105076446197927, + "learning_rate": 0.17649700465842078, + "loss": 0.0859, + "num_input_tokens_seen": 23003040, + "step": 17740 + }, + { + "epoch": 0.8670266044511763, + "grad_norm": 0.0009616142488084733, + "learning_rate": 0.17643902407337023, + "loss": 0.0927, + "num_input_tokens_seen": 23009440, + "step": 17745 + }, + { + "epoch": 0.8672709061148707, + "grad_norm": 0.0012631191639229655, + "learning_rate": 0.17638103941108993, + "loss": 0.0924, + "num_input_tokens_seen": 23015616, + "step": 17750 + }, + { + "epoch": 0.867515207778565, + "grad_norm": 0.0020245504565536976, + "learning_rate": 0.1763230506805218, + "loss": 0.1113, + "num_input_tokens_seen": 23021824, + "step": 17755 + }, + { + "epoch": 0.8677595094422593, + "grad_norm": 0.0009547359077259898, + "learning_rate": 0.1762650578906085, + "loss": 0.0975, + "num_input_tokens_seen": 23028448, + "step": 17760 + }, + { + "epoch": 0.8680038111059536, + "grad_norm": 0.0013802761677652597, + "learning_rate": 0.1762070610502932, + "loss": 0.1144, + "num_input_tokens_seen": 23035168, + "step": 17765 + }, + { + "epoch": 0.868248112769648, + "grad_norm": 0.0015353499911725521, + "learning_rate": 0.17614906016851975, + "loss": 0.1001, + "num_input_tokens_seen": 23041184, + "step": 17770 + }, + { + "epoch": 0.8684924144333422, + "grad_norm": 0.000657609780319035, + "learning_rate": 0.17609105525423258, + "loss": 0.0729, + "num_input_tokens_seen": 23047968, + "step": 17775 + }, + { + "epoch": 0.8687367160970366, + "grad_norm": 0.0009132391423918307, + "learning_rate": 0.1760330463163768, + "loss": 0.096, + "num_input_tokens_seen": 23054688, + "step": 17780 + }, + { + "epoch": 0.868981017760731, + "grad_norm": 0.0016734614036977291, + "learning_rate": 0.17597503336389816, + "loss": 0.0873, + "num_input_tokens_seen": 23060960, + "step": 17785 + }, + { + "epoch": 0.8692253194244253, + "grad_norm": 0.0015871862415224314, + "learning_rate": 0.17591701640574298, + "loss": 0.0905, + "num_input_tokens_seen": 23067648, + "step": 17790 + }, + { + "epoch": 0.8694696210881196, + "grad_norm": 0.0009661008953116834, + "learning_rate": 0.17585899545085815, + "loss": 0.1184, + "num_input_tokens_seen": 23074144, + "step": 17795 + }, + { + "epoch": 0.8697139227518139, + "grad_norm": 0.0009103638003580272, + "learning_rate": 0.17580097050819124, + "loss": 0.1053, + "num_input_tokens_seen": 23080448, + "step": 17800 + }, + { + "epoch": 0.8697139227518139, + "eval_loss": 0.10696081072092056, + "eval_runtime": 402.3622, + "eval_samples_per_second": 90.428, + "eval_steps_per_second": 22.609, + "num_input_tokens_seen": 23080448, + "step": 17800 + }, + { + "epoch": 0.8699582244155083, + "grad_norm": 0.0015655549941584468, + "learning_rate": 0.17574294158669046, + "loss": 0.1417, + "num_input_tokens_seen": 23086944, + "step": 17805 + }, + { + "epoch": 0.8702025260792026, + "grad_norm": 0.0007159115048125386, + "learning_rate": 0.17568490869530456, + "loss": 0.0935, + "num_input_tokens_seen": 23093280, + "step": 17810 + }, + { + "epoch": 0.870446827742897, + "grad_norm": 0.0007109255529940128, + "learning_rate": 0.17562687184298295, + "loss": 0.094, + "num_input_tokens_seen": 23099680, + "step": 17815 + }, + { + "epoch": 0.8706911294065912, + "grad_norm": 0.0005553108057938516, + "learning_rate": 0.1755688310386757, + "loss": 0.1026, + "num_input_tokens_seen": 23106080, + "step": 17820 + }, + { + "epoch": 0.8709354310702856, + "grad_norm": 0.0009056913550011814, + "learning_rate": 0.17551078629133335, + "loss": 0.0984, + "num_input_tokens_seen": 23112576, + "step": 17825 + }, + { + "epoch": 0.87117973273398, + "grad_norm": 0.0015032347291707993, + "learning_rate": 0.17545273760990718, + "loss": 0.0859, + "num_input_tokens_seen": 23118784, + "step": 17830 + }, + { + "epoch": 0.8714240343976742, + "grad_norm": 0.0015617478638887405, + "learning_rate": 0.17539468500334904, + "loss": 0.0768, + "num_input_tokens_seen": 23125504, + "step": 17835 + }, + { + "epoch": 0.8716683360613686, + "grad_norm": 0.0009047887288033962, + "learning_rate": 0.17533662848061132, + "loss": 0.0855, + "num_input_tokens_seen": 23132288, + "step": 17840 + }, + { + "epoch": 0.8719126377250629, + "grad_norm": 0.0009504143381491303, + "learning_rate": 0.1752785680506471, + "loss": 0.1011, + "num_input_tokens_seen": 23138592, + "step": 17845 + }, + { + "epoch": 0.8721569393887573, + "grad_norm": 0.0012771815527230501, + "learning_rate": 0.17522050372241, + "loss": 0.1107, + "num_input_tokens_seen": 23144960, + "step": 17850 + }, + { + "epoch": 0.8724012410524515, + "grad_norm": 0.0010689768241718411, + "learning_rate": 0.17516243550485425, + "loss": 0.1244, + "num_input_tokens_seen": 23151136, + "step": 17855 + }, + { + "epoch": 0.8726455427161459, + "grad_norm": 0.001067794975824654, + "learning_rate": 0.17510436340693478, + "loss": 0.0863, + "num_input_tokens_seen": 23157536, + "step": 17860 + }, + { + "epoch": 0.8728898443798402, + "grad_norm": 0.0013317321427166462, + "learning_rate": 0.175046287437607, + "loss": 0.0963, + "num_input_tokens_seen": 23163968, + "step": 17865 + }, + { + "epoch": 0.8731341460435346, + "grad_norm": 0.0014645860064774752, + "learning_rate": 0.17498820760582695, + "loss": 0.0899, + "num_input_tokens_seen": 23170688, + "step": 17870 + }, + { + "epoch": 0.8733784477072288, + "grad_norm": 0.0011482859263196588, + "learning_rate": 0.1749301239205512, + "loss": 0.1017, + "num_input_tokens_seen": 23177088, + "step": 17875 + }, + { + "epoch": 0.8736227493709232, + "grad_norm": 0.0009815521771088243, + "learning_rate": 0.1748720363907371, + "loss": 0.0959, + "num_input_tokens_seen": 23183360, + "step": 17880 + }, + { + "epoch": 0.8738670510346176, + "grad_norm": 0.0009367563179694116, + "learning_rate": 0.17481394502534242, + "loss": 0.0869, + "num_input_tokens_seen": 23189216, + "step": 17885 + }, + { + "epoch": 0.8741113526983119, + "grad_norm": 0.001539326156489551, + "learning_rate": 0.17475584983332562, + "loss": 0.067, + "num_input_tokens_seen": 23196704, + "step": 17890 + }, + { + "epoch": 0.8743556543620062, + "grad_norm": 0.0016654120991006494, + "learning_rate": 0.17469775082364558, + "loss": 0.1053, + "num_input_tokens_seen": 23202944, + "step": 17895 + }, + { + "epoch": 0.8745999560257005, + "grad_norm": 0.0006096675642766058, + "learning_rate": 0.17463964800526205, + "loss": 0.0811, + "num_input_tokens_seen": 23209184, + "step": 17900 + }, + { + "epoch": 0.8748442576893949, + "grad_norm": 0.0006020386936143041, + "learning_rate": 0.17458154138713522, + "loss": 0.0907, + "num_input_tokens_seen": 23215360, + "step": 17905 + }, + { + "epoch": 0.8750885593530892, + "grad_norm": 0.0013764443574473262, + "learning_rate": 0.17452343097822576, + "loss": 0.0946, + "num_input_tokens_seen": 23222048, + "step": 17910 + }, + { + "epoch": 0.8753328610167835, + "grad_norm": 0.002244947711005807, + "learning_rate": 0.17446531678749497, + "loss": 0.1117, + "num_input_tokens_seen": 23228608, + "step": 17915 + }, + { + "epoch": 0.8755771626804778, + "grad_norm": 0.0015071367379277945, + "learning_rate": 0.17440719882390496, + "loss": 0.0945, + "num_input_tokens_seen": 23235072, + "step": 17920 + }, + { + "epoch": 0.8758214643441722, + "grad_norm": 0.0015847309259697795, + "learning_rate": 0.17434907709641814, + "loss": 0.0937, + "num_input_tokens_seen": 23241472, + "step": 17925 + }, + { + "epoch": 0.8760657660078666, + "grad_norm": 0.00132970348931849, + "learning_rate": 0.17429095161399769, + "loss": 0.0794, + "num_input_tokens_seen": 23248160, + "step": 17930 + }, + { + "epoch": 0.8763100676715608, + "grad_norm": 0.0008747517131268978, + "learning_rate": 0.1742328223856072, + "loss": 0.0911, + "num_input_tokens_seen": 23254336, + "step": 17935 + }, + { + "epoch": 0.8765543693352552, + "grad_norm": 0.0011709468672052026, + "learning_rate": 0.174174689420211, + "loss": 0.0933, + "num_input_tokens_seen": 23260960, + "step": 17940 + }, + { + "epoch": 0.8767986709989495, + "grad_norm": 0.0014120095875114202, + "learning_rate": 0.1741165527267739, + "loss": 0.0803, + "num_input_tokens_seen": 23267488, + "step": 17945 + }, + { + "epoch": 0.8770429726626439, + "grad_norm": 0.0022734766826033592, + "learning_rate": 0.17405841231426125, + "loss": 0.094, + "num_input_tokens_seen": 23273824, + "step": 17950 + }, + { + "epoch": 0.8772872743263381, + "grad_norm": 0.001687565352767706, + "learning_rate": 0.1740002681916391, + "loss": 0.1015, + "num_input_tokens_seen": 23280384, + "step": 17955 + }, + { + "epoch": 0.8775315759900325, + "grad_norm": 0.0024535309057682753, + "learning_rate": 0.17394212036787401, + "loss": 0.1041, + "num_input_tokens_seen": 23286624, + "step": 17960 + }, + { + "epoch": 0.8777758776537268, + "grad_norm": 0.0014444037806242704, + "learning_rate": 0.1738839688519331, + "loss": 0.1293, + "num_input_tokens_seen": 23292736, + "step": 17965 + }, + { + "epoch": 0.8780201793174212, + "grad_norm": 0.0008229196537286043, + "learning_rate": 0.17382581365278402, + "loss": 0.0864, + "num_input_tokens_seen": 23299392, + "step": 17970 + }, + { + "epoch": 0.8782644809811155, + "grad_norm": 0.0013789385557174683, + "learning_rate": 0.17376765477939507, + "loss": 0.096, + "num_input_tokens_seen": 23305920, + "step": 17975 + }, + { + "epoch": 0.8785087826448098, + "grad_norm": 0.0013087056577205658, + "learning_rate": 0.1737094922407351, + "loss": 0.1011, + "num_input_tokens_seen": 23312416, + "step": 17980 + }, + { + "epoch": 0.8787530843085042, + "grad_norm": 0.0009320878889411688, + "learning_rate": 0.1736513260457734, + "loss": 0.0841, + "num_input_tokens_seen": 23318752, + "step": 17985 + }, + { + "epoch": 0.8789973859721985, + "grad_norm": 0.0010998154757544398, + "learning_rate": 0.17359315620348006, + "loss": 0.0964, + "num_input_tokens_seen": 23324960, + "step": 17990 + }, + { + "epoch": 0.8792416876358928, + "grad_norm": 0.0010202220873907208, + "learning_rate": 0.17353498272282547, + "loss": 0.1264, + "num_input_tokens_seen": 23331520, + "step": 17995 + }, + { + "epoch": 0.8794859892995871, + "grad_norm": 0.0012929122895002365, + "learning_rate": 0.17347680561278087, + "loss": 0.1144, + "num_input_tokens_seen": 23338016, + "step": 18000 + }, + { + "epoch": 0.8794859892995871, + "eval_loss": 0.09959163516759872, + "eval_runtime": 402.4541, + "eval_samples_per_second": 90.408, + "eval_steps_per_second": 22.604, + "num_input_tokens_seen": 23338016, + "step": 18000 + }, + { + "epoch": 0.8797302909632815, + "grad_norm": 0.0011279263999313116, + "learning_rate": 0.1734186248823178, + "loss": 0.0639, + "num_input_tokens_seen": 23344544, + "step": 18005 + }, + { + "epoch": 0.8799745926269757, + "grad_norm": 0.0010946077527478337, + "learning_rate": 0.17336044054040844, + "loss": 0.1003, + "num_input_tokens_seen": 23350848, + "step": 18010 + }, + { + "epoch": 0.8802188942906701, + "grad_norm": 0.0010508546838536859, + "learning_rate": 0.1733022525960256, + "loss": 0.1043, + "num_input_tokens_seen": 23357248, + "step": 18015 + }, + { + "epoch": 0.8804631959543644, + "grad_norm": 0.0013899615732952952, + "learning_rate": 0.1732440610581426, + "loss": 0.1217, + "num_input_tokens_seen": 23363584, + "step": 18020 + }, + { + "epoch": 0.8807074976180588, + "grad_norm": 0.0018810057081282139, + "learning_rate": 0.17318586593573326, + "loss": 0.1025, + "num_input_tokens_seen": 23369632, + "step": 18025 + }, + { + "epoch": 0.8809517992817532, + "grad_norm": 0.0011916294461116195, + "learning_rate": 0.17312766723777204, + "loss": 0.0902, + "num_input_tokens_seen": 23376640, + "step": 18030 + }, + { + "epoch": 0.8811961009454474, + "grad_norm": 0.0011011670576408505, + "learning_rate": 0.1730694649732339, + "loss": 0.1052, + "num_input_tokens_seen": 23383424, + "step": 18035 + }, + { + "epoch": 0.8814404026091418, + "grad_norm": 0.0013099833158776164, + "learning_rate": 0.17301125915109428, + "loss": 0.1047, + "num_input_tokens_seen": 23389824, + "step": 18040 + }, + { + "epoch": 0.8816847042728361, + "grad_norm": 0.0013852074043825269, + "learning_rate": 0.17295304978032938, + "loss": 0.1069, + "num_input_tokens_seen": 23396928, + "step": 18045 + }, + { + "epoch": 0.8819290059365305, + "grad_norm": 0.0015652908477932215, + "learning_rate": 0.17289483686991577, + "loss": 0.1045, + "num_input_tokens_seen": 23403264, + "step": 18050 + }, + { + "epoch": 0.8821733076002247, + "grad_norm": 0.0005831372691318393, + "learning_rate": 0.1728366204288306, + "loss": 0.0933, + "num_input_tokens_seen": 23410784, + "step": 18055 + }, + { + "epoch": 0.8824176092639191, + "grad_norm": 0.0010911409044638276, + "learning_rate": 0.17277840046605153, + "loss": 0.0856, + "num_input_tokens_seen": 23416864, + "step": 18060 + }, + { + "epoch": 0.8826619109276134, + "grad_norm": 0.0015303426189348102, + "learning_rate": 0.17272017699055686, + "loss": 0.0947, + "num_input_tokens_seen": 23423392, + "step": 18065 + }, + { + "epoch": 0.8829062125913077, + "grad_norm": 0.001624386408366263, + "learning_rate": 0.17266195001132542, + "loss": 0.0941, + "num_input_tokens_seen": 23429632, + "step": 18070 + }, + { + "epoch": 0.8831505142550021, + "grad_norm": 0.0009730748133733869, + "learning_rate": 0.17260371953733647, + "loss": 0.0832, + "num_input_tokens_seen": 23436416, + "step": 18075 + }, + { + "epoch": 0.8833948159186964, + "grad_norm": 0.0023997060488909483, + "learning_rate": 0.1725454855775699, + "loss": 0.1158, + "num_input_tokens_seen": 23442240, + "step": 18080 + }, + { + "epoch": 0.8836391175823908, + "grad_norm": 0.0010486753890290856, + "learning_rate": 0.17248724814100616, + "loss": 0.1098, + "num_input_tokens_seen": 23448352, + "step": 18085 + }, + { + "epoch": 0.883883419246085, + "grad_norm": 0.0011081094853579998, + "learning_rate": 0.17242900723662619, + "loss": 0.0855, + "num_input_tokens_seen": 23454528, + "step": 18090 + }, + { + "epoch": 0.8841277209097794, + "grad_norm": 0.0012391236377879977, + "learning_rate": 0.1723707628734114, + "loss": 0.0978, + "num_input_tokens_seen": 23460672, + "step": 18095 + }, + { + "epoch": 0.8843720225734737, + "grad_norm": 0.0018566002836450934, + "learning_rate": 0.1723125150603438, + "loss": 0.1046, + "num_input_tokens_seen": 23467200, + "step": 18100 + }, + { + "epoch": 0.8846163242371681, + "grad_norm": 0.0008991436916403472, + "learning_rate": 0.1722542638064061, + "loss": 0.1038, + "num_input_tokens_seen": 23473760, + "step": 18105 + }, + { + "epoch": 0.8848606259008623, + "grad_norm": 0.000953846494667232, + "learning_rate": 0.17219600912058117, + "loss": 0.0772, + "num_input_tokens_seen": 23479968, + "step": 18110 + }, + { + "epoch": 0.8851049275645567, + "grad_norm": 0.0017055791104212403, + "learning_rate": 0.17213775101185272, + "loss": 0.156, + "num_input_tokens_seen": 23486272, + "step": 18115 + }, + { + "epoch": 0.885349229228251, + "grad_norm": 0.0012551384279504418, + "learning_rate": 0.17207948948920485, + "loss": 0.1053, + "num_input_tokens_seen": 23493312, + "step": 18120 + }, + { + "epoch": 0.8855935308919454, + "grad_norm": 0.000977993244305253, + "learning_rate": 0.17202122456162228, + "loss": 0.0898, + "num_input_tokens_seen": 23500128, + "step": 18125 + }, + { + "epoch": 0.8858378325556397, + "grad_norm": 0.0006320507964119315, + "learning_rate": 0.17196295623809013, + "loss": 0.1003, + "num_input_tokens_seen": 23506624, + "step": 18130 + }, + { + "epoch": 0.886082134219334, + "grad_norm": 0.0007378446171060205, + "learning_rate": 0.1719046845275941, + "loss": 0.1062, + "num_input_tokens_seen": 23512864, + "step": 18135 + }, + { + "epoch": 0.8863264358830284, + "grad_norm": 0.0008878114749677479, + "learning_rate": 0.17184640943912044, + "loss": 0.0834, + "num_input_tokens_seen": 23519232, + "step": 18140 + }, + { + "epoch": 0.8865707375467227, + "grad_norm": 0.0014783969381824136, + "learning_rate": 0.1717881309816559, + "loss": 0.0782, + "num_input_tokens_seen": 23526336, + "step": 18145 + }, + { + "epoch": 0.886815039210417, + "grad_norm": 0.0012976133730262518, + "learning_rate": 0.1717298491641878, + "loss": 0.108, + "num_input_tokens_seen": 23532992, + "step": 18150 + }, + { + "epoch": 0.8870593408741113, + "grad_norm": 0.001684239599853754, + "learning_rate": 0.17167156399570385, + "loss": 0.0858, + "num_input_tokens_seen": 23539360, + "step": 18155 + }, + { + "epoch": 0.8873036425378057, + "grad_norm": 0.0014226491330191493, + "learning_rate": 0.17161327548519242, + "loss": 0.0841, + "num_input_tokens_seen": 23545856, + "step": 18160 + }, + { + "epoch": 0.8875479442015, + "grad_norm": 0.0013080649077892303, + "learning_rate": 0.1715549836416423, + "loss": 0.1201, + "num_input_tokens_seen": 23552384, + "step": 18165 + }, + { + "epoch": 0.8877922458651943, + "grad_norm": 0.001537008793093264, + "learning_rate": 0.17149668847404279, + "loss": 0.0975, + "num_input_tokens_seen": 23558560, + "step": 18170 + }, + { + "epoch": 0.8880365475288887, + "grad_norm": 0.0013924895320087671, + "learning_rate": 0.1714383899913838, + "loss": 0.1139, + "num_input_tokens_seen": 23564928, + "step": 18175 + }, + { + "epoch": 0.888280849192583, + "grad_norm": 0.0009441432193852961, + "learning_rate": 0.17138008820265563, + "loss": 0.0977, + "num_input_tokens_seen": 23571296, + "step": 18180 + }, + { + "epoch": 0.8885251508562774, + "grad_norm": 0.0006519706803373992, + "learning_rate": 0.17132178311684917, + "loss": 0.0903, + "num_input_tokens_seen": 23578080, + "step": 18185 + }, + { + "epoch": 0.8887694525199716, + "grad_norm": 0.0010697091929614544, + "learning_rate": 0.1712634747429559, + "loss": 0.1145, + "num_input_tokens_seen": 23584672, + "step": 18190 + }, + { + "epoch": 0.889013754183666, + "grad_norm": 0.0015900355065241456, + "learning_rate": 0.17120516308996753, + "loss": 0.1062, + "num_input_tokens_seen": 23591712, + "step": 18195 + }, + { + "epoch": 0.8892580558473603, + "grad_norm": 0.002369736786931753, + "learning_rate": 0.17114684816687653, + "loss": 0.1452, + "num_input_tokens_seen": 23598208, + "step": 18200 + }, + { + "epoch": 0.8892580558473603, + "eval_loss": 0.10160231590270996, + "eval_runtime": 402.5177, + "eval_samples_per_second": 90.394, + "eval_steps_per_second": 22.6, + "num_input_tokens_seen": 23598208, + "step": 18200 + }, + { + "epoch": 0.8895023575110547, + "grad_norm": 0.0008762368815951049, + "learning_rate": 0.17108852998267585, + "loss": 0.1127, + "num_input_tokens_seen": 23604864, + "step": 18205 + }, + { + "epoch": 0.8897466591747489, + "grad_norm": 0.001197539852000773, + "learning_rate": 0.17103020854635878, + "loss": 0.1177, + "num_input_tokens_seen": 23610880, + "step": 18210 + }, + { + "epoch": 0.8899909608384433, + "grad_norm": 0.00113668292760849, + "learning_rate": 0.1709718838669193, + "loss": 0.1122, + "num_input_tokens_seen": 23617408, + "step": 18215 + }, + { + "epoch": 0.8902352625021377, + "grad_norm": 0.0006209296989254653, + "learning_rate": 0.17091355595335173, + "loss": 0.0878, + "num_input_tokens_seen": 23623488, + "step": 18220 + }, + { + "epoch": 0.890479564165832, + "grad_norm": 0.0010418748715892434, + "learning_rate": 0.17085522481465107, + "loss": 0.11, + "num_input_tokens_seen": 23630368, + "step": 18225 + }, + { + "epoch": 0.8907238658295263, + "grad_norm": 0.0013626663712784648, + "learning_rate": 0.17079689045981264, + "loss": 0.0923, + "num_input_tokens_seen": 23636800, + "step": 18230 + }, + { + "epoch": 0.8909681674932206, + "grad_norm": 0.0009178203181363642, + "learning_rate": 0.17073855289783238, + "loss": 0.1031, + "num_input_tokens_seen": 23643072, + "step": 18235 + }, + { + "epoch": 0.891212469156915, + "grad_norm": 0.0007534818141721189, + "learning_rate": 0.1706802121377066, + "loss": 0.1151, + "num_input_tokens_seen": 23649568, + "step": 18240 + }, + { + "epoch": 0.8914567708206093, + "grad_norm": 0.0007735618855804205, + "learning_rate": 0.17062186818843225, + "loss": 0.1052, + "num_input_tokens_seen": 23655712, + "step": 18245 + }, + { + "epoch": 0.8917010724843036, + "grad_norm": 0.001235499745234847, + "learning_rate": 0.17056352105900668, + "loss": 0.1112, + "num_input_tokens_seen": 23661920, + "step": 18250 + }, + { + "epoch": 0.8919453741479979, + "grad_norm": 0.0010103486711159348, + "learning_rate": 0.17050517075842772, + "loss": 0.0893, + "num_input_tokens_seen": 23668608, + "step": 18255 + }, + { + "epoch": 0.8921896758116923, + "grad_norm": 0.001974795013666153, + "learning_rate": 0.17044681729569375, + "loss": 0.0862, + "num_input_tokens_seen": 23675168, + "step": 18260 + }, + { + "epoch": 0.8924339774753866, + "grad_norm": 0.0008690205286256969, + "learning_rate": 0.17038846067980365, + "loss": 0.0988, + "num_input_tokens_seen": 23681344, + "step": 18265 + }, + { + "epoch": 0.8926782791390809, + "grad_norm": 0.0012253456516191363, + "learning_rate": 0.17033010091975664, + "loss": 0.1234, + "num_input_tokens_seen": 23688128, + "step": 18270 + }, + { + "epoch": 0.8929225808027753, + "grad_norm": 0.0011033577611669898, + "learning_rate": 0.17027173802455262, + "loss": 0.0777, + "num_input_tokens_seen": 23694560, + "step": 18275 + }, + { + "epoch": 0.8931668824664696, + "grad_norm": 0.0011190008372068405, + "learning_rate": 0.1702133720031918, + "loss": 0.0682, + "num_input_tokens_seen": 23700960, + "step": 18280 + }, + { + "epoch": 0.893411184130164, + "grad_norm": 0.0012904421892017126, + "learning_rate": 0.17015500286467503, + "loss": 0.0866, + "num_input_tokens_seen": 23707680, + "step": 18285 + }, + { + "epoch": 0.8936554857938582, + "grad_norm": 0.000955405062995851, + "learning_rate": 0.17009663061800354, + "loss": 0.1014, + "num_input_tokens_seen": 23714720, + "step": 18290 + }, + { + "epoch": 0.8938997874575526, + "grad_norm": 0.001053531188517809, + "learning_rate": 0.17003825527217903, + "loss": 0.0763, + "num_input_tokens_seen": 23721088, + "step": 18295 + }, + { + "epoch": 0.8941440891212469, + "grad_norm": 0.0011657730210572481, + "learning_rate": 0.16997987683620377, + "loss": 0.0726, + "num_input_tokens_seen": 23727840, + "step": 18300 + }, + { + "epoch": 0.8943883907849413, + "grad_norm": 0.0007715621613897383, + "learning_rate": 0.16992149531908043, + "loss": 0.0853, + "num_input_tokens_seen": 23734304, + "step": 18305 + }, + { + "epoch": 0.8946326924486355, + "grad_norm": 0.0017138451803475618, + "learning_rate": 0.16986311072981214, + "loss": 0.1419, + "num_input_tokens_seen": 23740864, + "step": 18310 + }, + { + "epoch": 0.8948769941123299, + "grad_norm": 0.0021081191953271627, + "learning_rate": 0.16980472307740255, + "loss": 0.1008, + "num_input_tokens_seen": 23747328, + "step": 18315 + }, + { + "epoch": 0.8951212957760243, + "grad_norm": 0.0013997995993122458, + "learning_rate": 0.1697463323708558, + "loss": 0.0878, + "num_input_tokens_seen": 23754176, + "step": 18320 + }, + { + "epoch": 0.8953655974397186, + "grad_norm": 0.0014837788185104728, + "learning_rate": 0.16968793861917641, + "loss": 0.1231, + "num_input_tokens_seen": 23760160, + "step": 18325 + }, + { + "epoch": 0.8956098991034129, + "grad_norm": 0.0009548875386826694, + "learning_rate": 0.16962954183136952, + "loss": 0.0834, + "num_input_tokens_seen": 23766848, + "step": 18330 + }, + { + "epoch": 0.8958542007671072, + "grad_norm": 0.000676716910675168, + "learning_rate": 0.16957114201644058, + "loss": 0.0811, + "num_input_tokens_seen": 23773056, + "step": 18335 + }, + { + "epoch": 0.8960985024308016, + "grad_norm": 0.0015272983582690358, + "learning_rate": 0.16951273918339563, + "loss": 0.0957, + "num_input_tokens_seen": 23779584, + "step": 18340 + }, + { + "epoch": 0.8963428040944958, + "grad_norm": 0.001067801029421389, + "learning_rate": 0.16945433334124105, + "loss": 0.1036, + "num_input_tokens_seen": 23785728, + "step": 18345 + }, + { + "epoch": 0.8965871057581902, + "grad_norm": 0.0013169593876227736, + "learning_rate": 0.1693959244989838, + "loss": 0.0847, + "num_input_tokens_seen": 23792096, + "step": 18350 + }, + { + "epoch": 0.8968314074218845, + "grad_norm": 0.000960432575084269, + "learning_rate": 0.16933751266563127, + "loss": 0.0847, + "num_input_tokens_seen": 23798944, + "step": 18355 + }, + { + "epoch": 0.8970757090855789, + "grad_norm": 0.0010785760823637247, + "learning_rate": 0.16927909785019118, + "loss": 0.0964, + "num_input_tokens_seen": 23805120, + "step": 18360 + }, + { + "epoch": 0.8973200107492733, + "grad_norm": 0.0011734048603102565, + "learning_rate": 0.169220680061672, + "loss": 0.1085, + "num_input_tokens_seen": 23811424, + "step": 18365 + }, + { + "epoch": 0.8975643124129675, + "grad_norm": 0.0009492417448200285, + "learning_rate": 0.16916225930908244, + "loss": 0.0934, + "num_input_tokens_seen": 23818016, + "step": 18370 + }, + { + "epoch": 0.8978086140766619, + "grad_norm": 0.0008501993725076318, + "learning_rate": 0.16910383560143163, + "loss": 0.1034, + "num_input_tokens_seen": 23824512, + "step": 18375 + }, + { + "epoch": 0.8980529157403562, + "grad_norm": 0.0012351979967206717, + "learning_rate": 0.16904540894772935, + "loss": 0.1107, + "num_input_tokens_seen": 23830784, + "step": 18380 + }, + { + "epoch": 0.8982972174040506, + "grad_norm": 0.0006770442123524845, + "learning_rate": 0.16898697935698562, + "loss": 0.0704, + "num_input_tokens_seen": 23837888, + "step": 18385 + }, + { + "epoch": 0.8985415190677448, + "grad_norm": 0.0013293097727000713, + "learning_rate": 0.1689285468382111, + "loss": 0.1197, + "num_input_tokens_seen": 23844416, + "step": 18390 + }, + { + "epoch": 0.8987858207314392, + "grad_norm": 0.0013100142823532224, + "learning_rate": 0.16887011140041677, + "loss": 0.1141, + "num_input_tokens_seen": 23851296, + "step": 18395 + }, + { + "epoch": 0.8990301223951335, + "grad_norm": 0.0011508857132866979, + "learning_rate": 0.1688116730526141, + "loss": 0.1016, + "num_input_tokens_seen": 23857824, + "step": 18400 + }, + { + "epoch": 0.8990301223951335, + "eval_loss": 0.09983262419700623, + "eval_runtime": 402.1121, + "eval_samples_per_second": 90.485, + "eval_steps_per_second": 22.623, + "num_input_tokens_seen": 23857824, + "step": 18400 + }, + { + "epoch": 0.8992744240588278, + "grad_norm": 0.0007590417517349124, + "learning_rate": 0.1687532318038151, + "loss": 0.1072, + "num_input_tokens_seen": 23864480, + "step": 18405 + }, + { + "epoch": 0.8995187257225221, + "grad_norm": 0.0012915791012346745, + "learning_rate": 0.16869478766303206, + "loss": 0.0964, + "num_input_tokens_seen": 23871104, + "step": 18410 + }, + { + "epoch": 0.8997630273862165, + "grad_norm": 0.001271946239285171, + "learning_rate": 0.16863634063927788, + "loss": 0.085, + "num_input_tokens_seen": 23877440, + "step": 18415 + }, + { + "epoch": 0.9000073290499109, + "grad_norm": 0.0017389596905559301, + "learning_rate": 0.16857789074156568, + "loss": 0.1214, + "num_input_tokens_seen": 23883776, + "step": 18420 + }, + { + "epoch": 0.9002516307136051, + "grad_norm": 0.0015048555796965957, + "learning_rate": 0.16851943797890928, + "loss": 0.1071, + "num_input_tokens_seen": 23890400, + "step": 18425 + }, + { + "epoch": 0.9004959323772995, + "grad_norm": 0.0018800939433276653, + "learning_rate": 0.16846098236032284, + "loss": 0.1001, + "num_input_tokens_seen": 23896832, + "step": 18430 + }, + { + "epoch": 0.9007402340409938, + "grad_norm": 0.000807227217592299, + "learning_rate": 0.16840252389482097, + "loss": 0.1211, + "num_input_tokens_seen": 23903360, + "step": 18435 + }, + { + "epoch": 0.9009845357046882, + "grad_norm": 0.000938773388043046, + "learning_rate": 0.16834406259141857, + "loss": 0.085, + "num_input_tokens_seen": 23909984, + "step": 18440 + }, + { + "epoch": 0.9012288373683824, + "grad_norm": 0.0007601852412335575, + "learning_rate": 0.16828559845913124, + "loss": 0.1206, + "num_input_tokens_seen": 23916800, + "step": 18445 + }, + { + "epoch": 0.9014731390320768, + "grad_norm": 0.0009642993682064116, + "learning_rate": 0.16822713150697488, + "loss": 0.1041, + "num_input_tokens_seen": 23923552, + "step": 18450 + }, + { + "epoch": 0.9017174406957711, + "grad_norm": 0.0010969404829666018, + "learning_rate": 0.16816866174396575, + "loss": 0.1003, + "num_input_tokens_seen": 23930176, + "step": 18455 + }, + { + "epoch": 0.9019617423594655, + "grad_norm": 0.002045145956799388, + "learning_rate": 0.16811018917912057, + "loss": 0.1247, + "num_input_tokens_seen": 23936768, + "step": 18460 + }, + { + "epoch": 0.9022060440231598, + "grad_norm": 0.0012334029888734221, + "learning_rate": 0.16805171382145673, + "loss": 0.1015, + "num_input_tokens_seen": 23943424, + "step": 18465 + }, + { + "epoch": 0.9024503456868541, + "grad_norm": 0.0009139530011452734, + "learning_rate": 0.16799323567999175, + "loss": 0.1218, + "num_input_tokens_seen": 23949376, + "step": 18470 + }, + { + "epoch": 0.9026946473505485, + "grad_norm": 0.0005146160256117582, + "learning_rate": 0.16793475476374367, + "loss": 0.0942, + "num_input_tokens_seen": 23956192, + "step": 18475 + }, + { + "epoch": 0.9029389490142428, + "grad_norm": 0.0012714475160464644, + "learning_rate": 0.1678762710817311, + "loss": 0.107, + "num_input_tokens_seen": 23962528, + "step": 18480 + }, + { + "epoch": 0.9031832506779371, + "grad_norm": 0.00115728541277349, + "learning_rate": 0.1678177846429728, + "loss": 0.1095, + "num_input_tokens_seen": 23969280, + "step": 18485 + }, + { + "epoch": 0.9034275523416314, + "grad_norm": 0.0008521664422005415, + "learning_rate": 0.16775929545648827, + "loss": 0.1024, + "num_input_tokens_seen": 23975712, + "step": 18490 + }, + { + "epoch": 0.9036718540053258, + "grad_norm": 0.001088260905817151, + "learning_rate": 0.16770080353129715, + "loss": 0.1059, + "num_input_tokens_seen": 23982048, + "step": 18495 + }, + { + "epoch": 0.9039161556690201, + "grad_norm": 0.0013333780225366354, + "learning_rate": 0.16764230887641968, + "loss": 0.1021, + "num_input_tokens_seen": 23988416, + "step": 18500 + }, + { + "epoch": 0.9041604573327144, + "grad_norm": 0.0005508504691533744, + "learning_rate": 0.1675838115008765, + "loss": 0.0828, + "num_input_tokens_seen": 23994624, + "step": 18505 + }, + { + "epoch": 0.9044047589964088, + "grad_norm": 0.0011111338390037417, + "learning_rate": 0.1675253114136886, + "loss": 0.1152, + "num_input_tokens_seen": 24001056, + "step": 18510 + }, + { + "epoch": 0.9046490606601031, + "grad_norm": 0.002378551522269845, + "learning_rate": 0.16746680862387747, + "loss": 0.1257, + "num_input_tokens_seen": 24007264, + "step": 18515 + }, + { + "epoch": 0.9048933623237975, + "grad_norm": 0.0012058666907250881, + "learning_rate": 0.16740830314046493, + "loss": 0.1089, + "num_input_tokens_seen": 24013728, + "step": 18520 + }, + { + "epoch": 0.9051376639874917, + "grad_norm": 0.0010852636769413948, + "learning_rate": 0.1673497949724733, + "loss": 0.1109, + "num_input_tokens_seen": 24020640, + "step": 18525 + }, + { + "epoch": 0.9053819656511861, + "grad_norm": 0.0006837748805992305, + "learning_rate": 0.16729128412892522, + "loss": 0.0774, + "num_input_tokens_seen": 24027488, + "step": 18530 + }, + { + "epoch": 0.9056262673148804, + "grad_norm": 0.0011184620670974255, + "learning_rate": 0.16723277061884384, + "loss": 0.1004, + "num_input_tokens_seen": 24034240, + "step": 18535 + }, + { + "epoch": 0.9058705689785748, + "grad_norm": 0.0008628554060123861, + "learning_rate": 0.16717425445125267, + "loss": 0.0997, + "num_input_tokens_seen": 24040832, + "step": 18540 + }, + { + "epoch": 0.906114870642269, + "grad_norm": 0.0012041711015626788, + "learning_rate": 0.16711573563517565, + "loss": 0.0927, + "num_input_tokens_seen": 24047072, + "step": 18545 + }, + { + "epoch": 0.9063591723059634, + "grad_norm": 0.0009369706967845559, + "learning_rate": 0.1670572141796371, + "loss": 0.0893, + "num_input_tokens_seen": 24053600, + "step": 18550 + }, + { + "epoch": 0.9066034739696577, + "grad_norm": 0.0009175123414024711, + "learning_rate": 0.16699869009366175, + "loss": 0.093, + "num_input_tokens_seen": 24059904, + "step": 18555 + }, + { + "epoch": 0.9068477756333521, + "grad_norm": 0.0014569754712283611, + "learning_rate": 0.1669401633862748, + "loss": 0.103, + "num_input_tokens_seen": 24066112, + "step": 18560 + }, + { + "epoch": 0.9070920772970464, + "grad_norm": 0.001200773403979838, + "learning_rate": 0.16688163406650178, + "loss": 0.0867, + "num_input_tokens_seen": 24072416, + "step": 18565 + }, + { + "epoch": 0.9073363789607407, + "grad_norm": 0.000730053405277431, + "learning_rate": 0.1668231021433686, + "loss": 0.0794, + "num_input_tokens_seen": 24079168, + "step": 18570 + }, + { + "epoch": 0.9075806806244351, + "grad_norm": 0.000557315070182085, + "learning_rate": 0.1667645676259017, + "loss": 0.096, + "num_input_tokens_seen": 24085632, + "step": 18575 + }, + { + "epoch": 0.9078249822881294, + "grad_norm": 0.001066509634256363, + "learning_rate": 0.1667060305231277, + "loss": 0.1058, + "num_input_tokens_seen": 24091808, + "step": 18580 + }, + { + "epoch": 0.9080692839518237, + "grad_norm": 0.001567376428283751, + "learning_rate": 0.16664749084407396, + "loss": 0.094, + "num_input_tokens_seen": 24098208, + "step": 18585 + }, + { + "epoch": 0.908313585615518, + "grad_norm": 0.0012475622352212667, + "learning_rate": 0.16658894859776788, + "loss": 0.101, + "num_input_tokens_seen": 24104352, + "step": 18590 + }, + { + "epoch": 0.9085578872792124, + "grad_norm": 0.0018954958068206906, + "learning_rate": 0.16653040379323752, + "loss": 0.102, + "num_input_tokens_seen": 24110496, + "step": 18595 + }, + { + "epoch": 0.9088021889429067, + "grad_norm": 0.0011744211660698056, + "learning_rate": 0.16647185643951107, + "loss": 0.069, + "num_input_tokens_seen": 24117056, + "step": 18600 + }, + { + "epoch": 0.9088021889429067, + "eval_loss": 0.10439510643482208, + "eval_runtime": 402.3846, + "eval_samples_per_second": 90.423, + "eval_steps_per_second": 22.608, + "num_input_tokens_seen": 24117056, + "step": 18600 + }, + { + "epoch": 0.909046490606601, + "grad_norm": 0.001733164768666029, + "learning_rate": 0.1664133065456174, + "loss": 0.1177, + "num_input_tokens_seen": 24123392, + "step": 18605 + }, + { + "epoch": 0.9092907922702954, + "grad_norm": 0.0011774232843890786, + "learning_rate": 0.1663547541205856, + "loss": 0.1016, + "num_input_tokens_seen": 24129472, + "step": 18610 + }, + { + "epoch": 0.9095350939339897, + "grad_norm": 0.0005498698446899652, + "learning_rate": 0.16629619917344518, + "loss": 0.078, + "num_input_tokens_seen": 24136128, + "step": 18615 + }, + { + "epoch": 0.9097793955976841, + "grad_norm": 0.0006721717654727399, + "learning_rate": 0.16623764171322605, + "loss": 0.1254, + "num_input_tokens_seen": 24142848, + "step": 18620 + }, + { + "epoch": 0.9100236972613783, + "grad_norm": 0.0014413867611438036, + "learning_rate": 0.1661790817489585, + "loss": 0.0883, + "num_input_tokens_seen": 24149312, + "step": 18625 + }, + { + "epoch": 0.9102679989250727, + "grad_norm": 0.0014537929091602564, + "learning_rate": 0.16612051928967328, + "loss": 0.108, + "num_input_tokens_seen": 24155520, + "step": 18630 + }, + { + "epoch": 0.910512300588767, + "grad_norm": 0.0010093575110659003, + "learning_rate": 0.16606195434440138, + "loss": 0.1066, + "num_input_tokens_seen": 24162592, + "step": 18635 + }, + { + "epoch": 0.9107566022524614, + "grad_norm": 0.0012564202770590782, + "learning_rate": 0.16600338692217426, + "loss": 0.1113, + "num_input_tokens_seen": 24169120, + "step": 18640 + }, + { + "epoch": 0.9110009039161556, + "grad_norm": 0.0008949192124418914, + "learning_rate": 0.16594481703202374, + "loss": 0.0856, + "num_input_tokens_seen": 24175904, + "step": 18645 + }, + { + "epoch": 0.91124520557985, + "grad_norm": 0.0017049970338121057, + "learning_rate": 0.1658862446829821, + "loss": 0.1032, + "num_input_tokens_seen": 24182400, + "step": 18650 + }, + { + "epoch": 0.9114895072435444, + "grad_norm": 0.00099288416095078, + "learning_rate": 0.16582766988408187, + "loss": 0.1034, + "num_input_tokens_seen": 24188768, + "step": 18655 + }, + { + "epoch": 0.9117338089072387, + "grad_norm": 0.0012080238666385412, + "learning_rate": 0.16576909264435608, + "loss": 0.1071, + "num_input_tokens_seen": 24195200, + "step": 18660 + }, + { + "epoch": 0.911978110570933, + "grad_norm": 0.0007966308039613068, + "learning_rate": 0.16571051297283798, + "loss": 0.0793, + "num_input_tokens_seen": 24201792, + "step": 18665 + }, + { + "epoch": 0.9122224122346273, + "grad_norm": 0.0012484211474657059, + "learning_rate": 0.16565193087856137, + "loss": 0.1054, + "num_input_tokens_seen": 24208384, + "step": 18670 + }, + { + "epoch": 0.9124667138983217, + "grad_norm": 0.001752287382259965, + "learning_rate": 0.16559334637056033, + "loss": 0.0788, + "num_input_tokens_seen": 24214656, + "step": 18675 + }, + { + "epoch": 0.912711015562016, + "grad_norm": 0.0013693239307031035, + "learning_rate": 0.16553475945786933, + "loss": 0.1085, + "num_input_tokens_seen": 24221056, + "step": 18680 + }, + { + "epoch": 0.9129553172257103, + "grad_norm": 0.0011584892636165023, + "learning_rate": 0.16547617014952318, + "loss": 0.1063, + "num_input_tokens_seen": 24227200, + "step": 18685 + }, + { + "epoch": 0.9131996188894046, + "grad_norm": 0.0019367756322026253, + "learning_rate": 0.1654175784545571, + "loss": 0.1333, + "num_input_tokens_seen": 24233184, + "step": 18690 + }, + { + "epoch": 0.913443920553099, + "grad_norm": 0.0012163162464275956, + "learning_rate": 0.1653589843820067, + "loss": 0.0906, + "num_input_tokens_seen": 24239488, + "step": 18695 + }, + { + "epoch": 0.9136882222167932, + "grad_norm": 0.0010216563241556287, + "learning_rate": 0.1653003879409079, + "loss": 0.1153, + "num_input_tokens_seen": 24246048, + "step": 18700 + }, + { + "epoch": 0.9139325238804876, + "grad_norm": 0.0010831205872818828, + "learning_rate": 0.165241789140297, + "loss": 0.0816, + "num_input_tokens_seen": 24252256, + "step": 18705 + }, + { + "epoch": 0.914176825544182, + "grad_norm": 0.0008626091876067221, + "learning_rate": 0.16518318798921064, + "loss": 0.1062, + "num_input_tokens_seen": 24258528, + "step": 18710 + }, + { + "epoch": 0.9144211272078763, + "grad_norm": 0.0011942876735702157, + "learning_rate": 0.16512458449668593, + "loss": 0.0926, + "num_input_tokens_seen": 24264736, + "step": 18715 + }, + { + "epoch": 0.9146654288715707, + "grad_norm": 0.0009027039050124586, + "learning_rate": 0.1650659786717602, + "loss": 0.0965, + "num_input_tokens_seen": 24271584, + "step": 18720 + }, + { + "epoch": 0.9149097305352649, + "grad_norm": 0.0008794061723165214, + "learning_rate": 0.1650073705234712, + "loss": 0.1001, + "num_input_tokens_seen": 24278432, + "step": 18725 + }, + { + "epoch": 0.9151540321989593, + "grad_norm": 0.0011057850206270814, + "learning_rate": 0.16494876006085712, + "loss": 0.0974, + "num_input_tokens_seen": 24284704, + "step": 18730 + }, + { + "epoch": 0.9153983338626536, + "grad_norm": 0.001029975013807416, + "learning_rate": 0.16489014729295634, + "loss": 0.0935, + "num_input_tokens_seen": 24291392, + "step": 18735 + }, + { + "epoch": 0.915642635526348, + "grad_norm": 0.0007779513252899051, + "learning_rate": 0.16483153222880775, + "loss": 0.0946, + "num_input_tokens_seen": 24298176, + "step": 18740 + }, + { + "epoch": 0.9158869371900422, + "grad_norm": 0.0013535312609747052, + "learning_rate": 0.16477291487745052, + "loss": 0.1213, + "num_input_tokens_seen": 24304640, + "step": 18745 + }, + { + "epoch": 0.9161312388537366, + "grad_norm": 0.0008508235332556069, + "learning_rate": 0.16471429524792416, + "loss": 0.0764, + "num_input_tokens_seen": 24310816, + "step": 18750 + }, + { + "epoch": 0.916375540517431, + "grad_norm": 0.001121644745580852, + "learning_rate": 0.16465567334926856, + "loss": 0.0927, + "num_input_tokens_seen": 24317120, + "step": 18755 + }, + { + "epoch": 0.9166198421811252, + "grad_norm": 0.0009416216635145247, + "learning_rate": 0.16459704919052395, + "loss": 0.0969, + "num_input_tokens_seen": 24323648, + "step": 18760 + }, + { + "epoch": 0.9168641438448196, + "grad_norm": 0.0011642342433333397, + "learning_rate": 0.16453842278073086, + "loss": 0.0851, + "num_input_tokens_seen": 24330848, + "step": 18765 + }, + { + "epoch": 0.9171084455085139, + "grad_norm": 0.0010963613167405128, + "learning_rate": 0.16447979412893038, + "loss": 0.1019, + "num_input_tokens_seen": 24337312, + "step": 18770 + }, + { + "epoch": 0.9173527471722083, + "grad_norm": 0.0011319672921672463, + "learning_rate": 0.16442116324416367, + "loss": 0.1112, + "num_input_tokens_seen": 24343904, + "step": 18775 + }, + { + "epoch": 0.9175970488359025, + "grad_norm": 0.0014870504382997751, + "learning_rate": 0.1643625301354723, + "loss": 0.1035, + "num_input_tokens_seen": 24350496, + "step": 18780 + }, + { + "epoch": 0.9178413504995969, + "grad_norm": 0.0011122416472062469, + "learning_rate": 0.16430389481189828, + "loss": 0.0971, + "num_input_tokens_seen": 24356512, + "step": 18785 + }, + { + "epoch": 0.9180856521632912, + "grad_norm": 0.0010085884714499116, + "learning_rate": 0.164245257282484, + "loss": 0.0849, + "num_input_tokens_seen": 24362688, + "step": 18790 + }, + { + "epoch": 0.9183299538269856, + "grad_norm": 0.0011461536632850766, + "learning_rate": 0.16418661755627195, + "loss": 0.0647, + "num_input_tokens_seen": 24368992, + "step": 18795 + }, + { + "epoch": 0.9185742554906798, + "grad_norm": 0.0012559866299852729, + "learning_rate": 0.16412797564230527, + "loss": 0.0994, + "num_input_tokens_seen": 24375456, + "step": 18800 + }, + { + "epoch": 0.9185742554906798, + "eval_loss": 0.09842001646757126, + "eval_runtime": 402.0557, + "eval_samples_per_second": 90.497, + "eval_steps_per_second": 22.626, + "num_input_tokens_seen": 24375456, + "step": 18800 + }, + { + "epoch": 0.9188185571543742, + "grad_norm": 0.0013574231415987015, + "learning_rate": 0.16406933154962713, + "loss": 0.0758, + "num_input_tokens_seen": 24381888, + "step": 18805 + }, + { + "epoch": 0.9190628588180686, + "grad_norm": 0.0015960369491949677, + "learning_rate": 0.16401068528728133, + "loss": 0.0712, + "num_input_tokens_seen": 24388384, + "step": 18810 + }, + { + "epoch": 0.9193071604817629, + "grad_norm": 0.0014185927575454116, + "learning_rate": 0.16395203686431173, + "loss": 0.1011, + "num_input_tokens_seen": 24394880, + "step": 18815 + }, + { + "epoch": 0.9195514621454572, + "grad_norm": 0.0016832206165418029, + "learning_rate": 0.16389338628976277, + "loss": 0.0781, + "num_input_tokens_seen": 24401184, + "step": 18820 + }, + { + "epoch": 0.9197957638091515, + "grad_norm": 0.0017738890601322055, + "learning_rate": 0.163834733572679, + "loss": 0.1085, + "num_input_tokens_seen": 24407520, + "step": 18825 + }, + { + "epoch": 0.9200400654728459, + "grad_norm": 0.0010168743319809437, + "learning_rate": 0.16377607872210545, + "loss": 0.129, + "num_input_tokens_seen": 24414080, + "step": 18830 + }, + { + "epoch": 0.9202843671365402, + "grad_norm": 0.0013779945438727736, + "learning_rate": 0.16371742174708748, + "loss": 0.1037, + "num_input_tokens_seen": 24420224, + "step": 18835 + }, + { + "epoch": 0.9205286688002345, + "grad_norm": 0.0014031052123755217, + "learning_rate": 0.16365876265667065, + "loss": 0.1034, + "num_input_tokens_seen": 24426304, + "step": 18840 + }, + { + "epoch": 0.9207729704639288, + "grad_norm": 0.0007120507070794702, + "learning_rate": 0.163600101459901, + "loss": 0.0983, + "num_input_tokens_seen": 24433216, + "step": 18845 + }, + { + "epoch": 0.9210172721276232, + "grad_norm": 0.0012914633844047785, + "learning_rate": 0.16354143816582484, + "loss": 0.0847, + "num_input_tokens_seen": 24440000, + "step": 18850 + }, + { + "epoch": 0.9212615737913176, + "grad_norm": 0.0016128761926665902, + "learning_rate": 0.1634827727834887, + "loss": 0.1033, + "num_input_tokens_seen": 24446496, + "step": 18855 + }, + { + "epoch": 0.9215058754550118, + "grad_norm": 0.0017223170725628734, + "learning_rate": 0.16342410532193954, + "loss": 0.1076, + "num_input_tokens_seen": 24453088, + "step": 18860 + }, + { + "epoch": 0.9217501771187062, + "grad_norm": 0.0013406621292233467, + "learning_rate": 0.16336543579022464, + "loss": 0.114, + "num_input_tokens_seen": 24460128, + "step": 18865 + }, + { + "epoch": 0.9219944787824005, + "grad_norm": 0.0013527476694434881, + "learning_rate": 0.16330676419739157, + "loss": 0.0896, + "num_input_tokens_seen": 24466336, + "step": 18870 + }, + { + "epoch": 0.9222387804460949, + "grad_norm": 0.0015047970227897167, + "learning_rate": 0.1632480905524883, + "loss": 0.1067, + "num_input_tokens_seen": 24472672, + "step": 18875 + }, + { + "epoch": 0.9224830821097891, + "grad_norm": 0.0011389334686100483, + "learning_rate": 0.16318941486456293, + "loss": 0.0912, + "num_input_tokens_seen": 24479488, + "step": 18880 + }, + { + "epoch": 0.9227273837734835, + "grad_norm": 0.0011636913986876607, + "learning_rate": 0.16313073714266405, + "loss": 0.1024, + "num_input_tokens_seen": 24485632, + "step": 18885 + }, + { + "epoch": 0.9229716854371778, + "grad_norm": 0.0012018232373520732, + "learning_rate": 0.16307205739584052, + "loss": 0.0939, + "num_input_tokens_seen": 24492224, + "step": 18890 + }, + { + "epoch": 0.9232159871008722, + "grad_norm": 0.0014246703358367085, + "learning_rate": 0.16301337563314144, + "loss": 0.1095, + "num_input_tokens_seen": 24499008, + "step": 18895 + }, + { + "epoch": 0.9234602887645665, + "grad_norm": 0.0013513892190530896, + "learning_rate": 0.1629546918636163, + "loss": 0.0836, + "num_input_tokens_seen": 24505568, + "step": 18900 + }, + { + "epoch": 0.9237045904282608, + "grad_norm": 0.0013390325475484133, + "learning_rate": 0.16289600609631485, + "loss": 0.0853, + "num_input_tokens_seen": 24512224, + "step": 18905 + }, + { + "epoch": 0.9239488920919552, + "grad_norm": 0.0018386553274467587, + "learning_rate": 0.16283731834028722, + "loss": 0.0958, + "num_input_tokens_seen": 24518944, + "step": 18910 + }, + { + "epoch": 0.9241931937556495, + "grad_norm": 0.001403379370458424, + "learning_rate": 0.16277862860458378, + "loss": 0.0823, + "num_input_tokens_seen": 24525408, + "step": 18915 + }, + { + "epoch": 0.9244374954193438, + "grad_norm": 0.0010635392973199487, + "learning_rate": 0.16271993689825526, + "loss": 0.0725, + "num_input_tokens_seen": 24531712, + "step": 18920 + }, + { + "epoch": 0.9246817970830381, + "grad_norm": 0.0020055552013218403, + "learning_rate": 0.1626612432303526, + "loss": 0.0974, + "num_input_tokens_seen": 24538304, + "step": 18925 + }, + { + "epoch": 0.9249260987467325, + "grad_norm": 0.0013316269032657146, + "learning_rate": 0.1626025476099271, + "loss": 0.1023, + "num_input_tokens_seen": 24544352, + "step": 18930 + }, + { + "epoch": 0.9251704004104268, + "grad_norm": 0.003679343033581972, + "learning_rate": 0.1625438500460304, + "loss": 0.1129, + "num_input_tokens_seen": 24551168, + "step": 18935 + }, + { + "epoch": 0.9254147020741211, + "grad_norm": 0.0010551285231485963, + "learning_rate": 0.16248515054771442, + "loss": 0.0855, + "num_input_tokens_seen": 24557536, + "step": 18940 + }, + { + "epoch": 0.9256590037378154, + "grad_norm": 0.0011512935161590576, + "learning_rate": 0.16242644912403123, + "loss": 0.1326, + "num_input_tokens_seen": 24563680, + "step": 18945 + }, + { + "epoch": 0.9259033054015098, + "grad_norm": 0.0014117697719484568, + "learning_rate": 0.1623677457840335, + "loss": 0.1063, + "num_input_tokens_seen": 24570336, + "step": 18950 + }, + { + "epoch": 0.9261476070652042, + "grad_norm": 0.0017275229329243302, + "learning_rate": 0.16230904053677397, + "loss": 0.1159, + "num_input_tokens_seen": 24576704, + "step": 18955 + }, + { + "epoch": 0.9263919087288984, + "grad_norm": 0.0009987277444452047, + "learning_rate": 0.16225033339130568, + "loss": 0.1073, + "num_input_tokens_seen": 24583072, + "step": 18960 + }, + { + "epoch": 0.9266362103925928, + "grad_norm": 0.0007755791302770376, + "learning_rate": 0.16219162435668197, + "loss": 0.0907, + "num_input_tokens_seen": 24589504, + "step": 18965 + }, + { + "epoch": 0.9268805120562871, + "grad_norm": 0.0013865954242646694, + "learning_rate": 0.16213291344195666, + "loss": 0.0879, + "num_input_tokens_seen": 24595936, + "step": 18970 + }, + { + "epoch": 0.9271248137199815, + "grad_norm": 0.0014186679618433118, + "learning_rate": 0.16207420065618358, + "loss": 0.1235, + "num_input_tokens_seen": 24602560, + "step": 18975 + }, + { + "epoch": 0.9273691153836757, + "grad_norm": 0.000515941355843097, + "learning_rate": 0.16201548600841706, + "loss": 0.0956, + "num_input_tokens_seen": 24609440, + "step": 18980 + }, + { + "epoch": 0.9276134170473701, + "grad_norm": 0.0008853126782923937, + "learning_rate": 0.16195676950771154, + "loss": 0.1167, + "num_input_tokens_seen": 24616384, + "step": 18985 + }, + { + "epoch": 0.9278577187110644, + "grad_norm": 0.001172257587313652, + "learning_rate": 0.16189805116312198, + "loss": 0.115, + "num_input_tokens_seen": 24622784, + "step": 18990 + }, + { + "epoch": 0.9281020203747588, + "grad_norm": 0.0009775747312232852, + "learning_rate": 0.16183933098370337, + "loss": 0.0956, + "num_input_tokens_seen": 24629344, + "step": 18995 + }, + { + "epoch": 0.9283463220384531, + "grad_norm": 0.0013818592997267842, + "learning_rate": 0.16178060897851115, + "loss": 0.0993, + "num_input_tokens_seen": 24635712, + "step": 19000 + }, + { + "epoch": 0.9283463220384531, + "eval_loss": 0.101136215031147, + "eval_runtime": 402.4716, + "eval_samples_per_second": 90.404, + "eval_steps_per_second": 22.603, + "num_input_tokens_seen": 24635712, + "step": 19000 + }, + { + "epoch": 0.9285906237021474, + "grad_norm": 0.0008406028500758111, + "learning_rate": 0.16172188515660096, + "loss": 0.0856, + "num_input_tokens_seen": 24642432, + "step": 19005 + }, + { + "epoch": 0.9288349253658418, + "grad_norm": 0.0010870688129216433, + "learning_rate": 0.16166315952702878, + "loss": 0.0953, + "num_input_tokens_seen": 24648896, + "step": 19010 + }, + { + "epoch": 0.929079227029536, + "grad_norm": 0.0009436120744794607, + "learning_rate": 0.16160443209885084, + "loss": 0.0831, + "num_input_tokens_seen": 24655360, + "step": 19015 + }, + { + "epoch": 0.9293235286932304, + "grad_norm": 0.0013168652076274157, + "learning_rate": 0.16154570288112363, + "loss": 0.0723, + "num_input_tokens_seen": 24662336, + "step": 19020 + }, + { + "epoch": 0.9295678303569247, + "grad_norm": 0.000862950284499675, + "learning_rate": 0.16148697188290395, + "loss": 0.0755, + "num_input_tokens_seen": 24668640, + "step": 19025 + }, + { + "epoch": 0.9298121320206191, + "grad_norm": 0.001860008342191577, + "learning_rate": 0.16142823911324888, + "loss": 0.0914, + "num_input_tokens_seen": 24675520, + "step": 19030 + }, + { + "epoch": 0.9300564336843133, + "grad_norm": 0.0018334456253796816, + "learning_rate": 0.16136950458121568, + "loss": 0.0909, + "num_input_tokens_seen": 24681632, + "step": 19035 + }, + { + "epoch": 0.9303007353480077, + "grad_norm": 0.0012097519356757402, + "learning_rate": 0.16131076829586205, + "loss": 0.0824, + "num_input_tokens_seen": 24688032, + "step": 19040 + }, + { + "epoch": 0.9305450370117021, + "grad_norm": 0.0013599824160337448, + "learning_rate": 0.1612520302662457, + "loss": 0.0801, + "num_input_tokens_seen": 24694208, + "step": 19045 + }, + { + "epoch": 0.9307893386753964, + "grad_norm": 0.0010122046805918217, + "learning_rate": 0.16119329050142497, + "loss": 0.0891, + "num_input_tokens_seen": 24701184, + "step": 19050 + }, + { + "epoch": 0.9310336403390908, + "grad_norm": 0.0021723138634115458, + "learning_rate": 0.16113454901045818, + "loss": 0.0982, + "num_input_tokens_seen": 24707616, + "step": 19055 + }, + { + "epoch": 0.931277942002785, + "grad_norm": 0.0007712012738920748, + "learning_rate": 0.16107580580240397, + "loss": 0.0883, + "num_input_tokens_seen": 24713984, + "step": 19060 + }, + { + "epoch": 0.9315222436664794, + "grad_norm": 0.0012977428268641233, + "learning_rate": 0.16101706088632134, + "loss": 0.082, + "num_input_tokens_seen": 24721024, + "step": 19065 + }, + { + "epoch": 0.9317665453301737, + "grad_norm": 0.0013577204663306475, + "learning_rate": 0.16095831427126947, + "loss": 0.0994, + "num_input_tokens_seen": 24727712, + "step": 19070 + }, + { + "epoch": 0.932010846993868, + "grad_norm": 0.0012060819426551461, + "learning_rate": 0.16089956596630783, + "loss": 0.1302, + "num_input_tokens_seen": 24733952, + "step": 19075 + }, + { + "epoch": 0.9322551486575623, + "grad_norm": 0.0011010312009602785, + "learning_rate": 0.16084081598049618, + "loss": 0.0849, + "num_input_tokens_seen": 24740576, + "step": 19080 + }, + { + "epoch": 0.9324994503212567, + "grad_norm": 0.0025496706366539, + "learning_rate": 0.1607820643228944, + "loss": 0.1017, + "num_input_tokens_seen": 24747264, + "step": 19085 + }, + { + "epoch": 0.932743751984951, + "grad_norm": 0.001058381050825119, + "learning_rate": 0.16072331100256285, + "loss": 0.0701, + "num_input_tokens_seen": 24753568, + "step": 19090 + }, + { + "epoch": 0.9329880536486453, + "grad_norm": 0.001296088332310319, + "learning_rate": 0.16066455602856197, + "loss": 0.1073, + "num_input_tokens_seen": 24760096, + "step": 19095 + }, + { + "epoch": 0.9332323553123397, + "grad_norm": 0.0011527694296091795, + "learning_rate": 0.16060579940995257, + "loss": 0.0914, + "num_input_tokens_seen": 24766240, + "step": 19100 + }, + { + "epoch": 0.933476656976034, + "grad_norm": 0.001625058357603848, + "learning_rate": 0.16054704115579557, + "loss": 0.1028, + "num_input_tokens_seen": 24772608, + "step": 19105 + }, + { + "epoch": 0.9337209586397284, + "grad_norm": 0.001108926604501903, + "learning_rate": 0.1604882812751523, + "loss": 0.0888, + "num_input_tokens_seen": 24779072, + "step": 19110 + }, + { + "epoch": 0.9339652603034226, + "grad_norm": 0.0009971476392820477, + "learning_rate": 0.16042951977708425, + "loss": 0.0782, + "num_input_tokens_seen": 24785792, + "step": 19115 + }, + { + "epoch": 0.934209561967117, + "grad_norm": 0.000931841554120183, + "learning_rate": 0.16037075667065318, + "loss": 0.083, + "num_input_tokens_seen": 24792320, + "step": 19120 + }, + { + "epoch": 0.9344538636308113, + "grad_norm": 0.0009178174077533185, + "learning_rate": 0.1603119919649211, + "loss": 0.0845, + "num_input_tokens_seen": 24798784, + "step": 19125 + }, + { + "epoch": 0.9346981652945057, + "grad_norm": 0.0027470497880131006, + "learning_rate": 0.16025322566895028, + "loss": 0.1154, + "num_input_tokens_seen": 24805408, + "step": 19130 + }, + { + "epoch": 0.9349424669581999, + "grad_norm": 0.0012623803922906518, + "learning_rate": 0.16019445779180322, + "loss": 0.1279, + "num_input_tokens_seen": 24811264, + "step": 19135 + }, + { + "epoch": 0.9351867686218943, + "grad_norm": 0.0012766923755407333, + "learning_rate": 0.16013568834254271, + "loss": 0.1084, + "num_input_tokens_seen": 24817536, + "step": 19140 + }, + { + "epoch": 0.9354310702855887, + "grad_norm": 0.0008311670389957726, + "learning_rate": 0.1600769173302316, + "loss": 0.0958, + "num_input_tokens_seen": 24823936, + "step": 19145 + }, + { + "epoch": 0.935675371949283, + "grad_norm": 0.0008147009648382664, + "learning_rate": 0.16001814476393322, + "loss": 0.105, + "num_input_tokens_seen": 24829792, + "step": 19150 + }, + { + "epoch": 0.9359196736129773, + "grad_norm": 0.0007778106955811381, + "learning_rate": 0.15995937065271104, + "loss": 0.093, + "num_input_tokens_seen": 24836448, + "step": 19155 + }, + { + "epoch": 0.9361639752766716, + "grad_norm": 0.0012062181485816836, + "learning_rate": 0.15990059500562873, + "loss": 0.1014, + "num_input_tokens_seen": 24843488, + "step": 19160 + }, + { + "epoch": 0.936408276940366, + "grad_norm": 0.00150828727055341, + "learning_rate": 0.15984181783175025, + "loss": 0.1079, + "num_input_tokens_seen": 24849568, + "step": 19165 + }, + { + "epoch": 0.9366525786040603, + "grad_norm": 0.0013758647255599499, + "learning_rate": 0.1597830391401398, + "loss": 0.0971, + "num_input_tokens_seen": 24856608, + "step": 19170 + }, + { + "epoch": 0.9368968802677546, + "grad_norm": 0.0005973263178020716, + "learning_rate": 0.15972425893986178, + "loss": 0.0863, + "num_input_tokens_seen": 24863136, + "step": 19175 + }, + { + "epoch": 0.9371411819314489, + "grad_norm": 0.0011935210786759853, + "learning_rate": 0.15966547723998084, + "loss": 0.1058, + "num_input_tokens_seen": 24869472, + "step": 19180 + }, + { + "epoch": 0.9373854835951433, + "grad_norm": 0.0007520189974457026, + "learning_rate": 0.15960669404956176, + "loss": 0.1052, + "num_input_tokens_seen": 24875584, + "step": 19185 + }, + { + "epoch": 0.9376297852588377, + "grad_norm": 0.0016606722492724657, + "learning_rate": 0.1595479093776698, + "loss": 0.0891, + "num_input_tokens_seen": 24881952, + "step": 19190 + }, + { + "epoch": 0.9378740869225319, + "grad_norm": 0.0011991179781034589, + "learning_rate": 0.15948912323337022, + "loss": 0.1021, + "num_input_tokens_seen": 24888928, + "step": 19195 + }, + { + "epoch": 0.9381183885862263, + "grad_norm": 0.002177310176193714, + "learning_rate": 0.1594303356257286, + "loss": 0.1135, + "num_input_tokens_seen": 24895360, + "step": 19200 + }, + { + "epoch": 0.9381183885862263, + "eval_loss": 0.10376187413930893, + "eval_runtime": 402.6184, + "eval_samples_per_second": 90.371, + "eval_steps_per_second": 22.595, + "num_input_tokens_seen": 24895360, + "step": 19200 + }, + { + "epoch": 0.9383626902499206, + "grad_norm": 0.00044654670637100935, + "learning_rate": 0.15937154656381072, + "loss": 0.1026, + "num_input_tokens_seen": 24902208, + "step": 19205 + }, + { + "epoch": 0.938606991913615, + "grad_norm": 0.0010098859202116728, + "learning_rate": 0.15931275605668258, + "loss": 0.07, + "num_input_tokens_seen": 24908576, + "step": 19210 + }, + { + "epoch": 0.9388512935773092, + "grad_norm": 0.0011857416247949004, + "learning_rate": 0.1592539641134104, + "loss": 0.1001, + "num_input_tokens_seen": 24914912, + "step": 19215 + }, + { + "epoch": 0.9390955952410036, + "grad_norm": 0.0006205565296113491, + "learning_rate": 0.1591951707430607, + "loss": 0.079, + "num_input_tokens_seen": 24921664, + "step": 19220 + }, + { + "epoch": 0.9393398969046979, + "grad_norm": 0.0012430199421942234, + "learning_rate": 0.15913637595470007, + "loss": 0.0957, + "num_input_tokens_seen": 24928576, + "step": 19225 + }, + { + "epoch": 0.9395841985683923, + "grad_norm": 0.0008997889817692339, + "learning_rate": 0.15907757975739548, + "loss": 0.0871, + "num_input_tokens_seen": 24934912, + "step": 19230 + }, + { + "epoch": 0.9398285002320865, + "grad_norm": 0.0007351747481152415, + "learning_rate": 0.159018782160214, + "loss": 0.0923, + "num_input_tokens_seen": 24940992, + "step": 19235 + }, + { + "epoch": 0.9400728018957809, + "grad_norm": 0.001309903571382165, + "learning_rate": 0.158959983172223, + "loss": 0.0753, + "num_input_tokens_seen": 24947968, + "step": 19240 + }, + { + "epoch": 0.9403171035594753, + "grad_norm": 0.004359635524451733, + "learning_rate": 0.15890118280249, + "loss": 0.1218, + "num_input_tokens_seen": 24954784, + "step": 19245 + }, + { + "epoch": 0.9405614052231696, + "grad_norm": 0.0006782159325666726, + "learning_rate": 0.15884238106008275, + "loss": 0.1035, + "num_input_tokens_seen": 24961440, + "step": 19250 + }, + { + "epoch": 0.9408057068868639, + "grad_norm": 0.0015973779372870922, + "learning_rate": 0.15878357795406922, + "loss": 0.0925, + "num_input_tokens_seen": 24967680, + "step": 19255 + }, + { + "epoch": 0.9410500085505582, + "grad_norm": 0.0006894150865264237, + "learning_rate": 0.15872477349351757, + "loss": 0.0801, + "num_input_tokens_seen": 24974368, + "step": 19260 + }, + { + "epoch": 0.9412943102142526, + "grad_norm": 0.001273287576623261, + "learning_rate": 0.15866596768749622, + "loss": 0.0853, + "num_input_tokens_seen": 24981056, + "step": 19265 + }, + { + "epoch": 0.9415386118779469, + "grad_norm": 0.001358149223960936, + "learning_rate": 0.15860716054507373, + "loss": 0.0929, + "num_input_tokens_seen": 24987424, + "step": 19270 + }, + { + "epoch": 0.9417829135416412, + "grad_norm": 0.0015591745031997561, + "learning_rate": 0.1585483520753189, + "loss": 0.1285, + "num_input_tokens_seen": 24994048, + "step": 19275 + }, + { + "epoch": 0.9420272152053355, + "grad_norm": 0.0011547757312655449, + "learning_rate": 0.1584895422873008, + "loss": 0.0963, + "num_input_tokens_seen": 25000736, + "step": 19280 + }, + { + "epoch": 0.9422715168690299, + "grad_norm": 0.0016491300193592906, + "learning_rate": 0.1584307311900886, + "loss": 0.1161, + "num_input_tokens_seen": 25007104, + "step": 19285 + }, + { + "epoch": 0.9425158185327243, + "grad_norm": 0.0009343376732431352, + "learning_rate": 0.1583719187927517, + "loss": 0.0526, + "num_input_tokens_seen": 25013760, + "step": 19290 + }, + { + "epoch": 0.9427601201964185, + "grad_norm": 0.0007753124227747321, + "learning_rate": 0.15831310510435967, + "loss": 0.11, + "num_input_tokens_seen": 25020672, + "step": 19295 + }, + { + "epoch": 0.9430044218601129, + "grad_norm": 0.0009972767438739538, + "learning_rate": 0.15825429013398243, + "loss": 0.0959, + "num_input_tokens_seen": 25027136, + "step": 19300 + }, + { + "epoch": 0.9432487235238072, + "grad_norm": 0.0008013320621103048, + "learning_rate": 0.15819547389068986, + "loss": 0.0785, + "num_input_tokens_seen": 25033696, + "step": 19305 + }, + { + "epoch": 0.9434930251875016, + "grad_norm": 0.0016149379080161452, + "learning_rate": 0.1581366563835522, + "loss": 0.1008, + "num_input_tokens_seen": 25040448, + "step": 19310 + }, + { + "epoch": 0.9437373268511958, + "grad_norm": 0.0011344965314492583, + "learning_rate": 0.15807783762163993, + "loss": 0.1242, + "num_input_tokens_seen": 25046976, + "step": 19315 + }, + { + "epoch": 0.9439816285148902, + "grad_norm": 0.0017693066038191319, + "learning_rate": 0.15801901761402365, + "loss": 0.1057, + "num_input_tokens_seen": 25053728, + "step": 19320 + }, + { + "epoch": 0.9442259301785845, + "grad_norm": 0.0006893172394484282, + "learning_rate": 0.157960196369774, + "loss": 0.0711, + "num_input_tokens_seen": 25060416, + "step": 19325 + }, + { + "epoch": 0.9444702318422789, + "grad_norm": 0.0012675028992816806, + "learning_rate": 0.157901373897962, + "loss": 0.1014, + "num_input_tokens_seen": 25066656, + "step": 19330 + }, + { + "epoch": 0.9447145335059732, + "grad_norm": 0.0013458088506013155, + "learning_rate": 0.15784255020765892, + "loss": 0.1043, + "num_input_tokens_seen": 25072768, + "step": 19335 + }, + { + "epoch": 0.9449588351696675, + "grad_norm": 0.001341189374215901, + "learning_rate": 0.157783725307936, + "loss": 0.1151, + "num_input_tokens_seen": 25079104, + "step": 19340 + }, + { + "epoch": 0.9452031368333619, + "grad_norm": 0.0011210946831852198, + "learning_rate": 0.15772489920786484, + "loss": 0.0791, + "num_input_tokens_seen": 25085984, + "step": 19345 + }, + { + "epoch": 0.9454474384970561, + "grad_norm": 0.0006888416828587651, + "learning_rate": 0.15766607191651713, + "loss": 0.1014, + "num_input_tokens_seen": 25092640, + "step": 19350 + }, + { + "epoch": 0.9456917401607505, + "grad_norm": 0.0008195184054784477, + "learning_rate": 0.1576072434429648, + "loss": 0.0719, + "num_input_tokens_seen": 25099136, + "step": 19355 + }, + { + "epoch": 0.9459360418244448, + "grad_norm": 0.0017577123362571, + "learning_rate": 0.15754841379627998, + "loss": 0.1001, + "num_input_tokens_seen": 25105600, + "step": 19360 + }, + { + "epoch": 0.9461803434881392, + "grad_norm": 0.001143959118053317, + "learning_rate": 0.15748958298553484, + "loss": 0.0836, + "num_input_tokens_seen": 25111872, + "step": 19365 + }, + { + "epoch": 0.9464246451518334, + "grad_norm": 0.0010538706555962563, + "learning_rate": 0.1574307510198019, + "loss": 0.0828, + "num_input_tokens_seen": 25118464, + "step": 19370 + }, + { + "epoch": 0.9466689468155278, + "grad_norm": 0.0010763087775558233, + "learning_rate": 0.15737191790815375, + "loss": 0.1158, + "num_input_tokens_seen": 25124512, + "step": 19375 + }, + { + "epoch": 0.9469132484792221, + "grad_norm": 0.001351845101453364, + "learning_rate": 0.15731308365966323, + "loss": 0.0837, + "num_input_tokens_seen": 25130752, + "step": 19380 + }, + { + "epoch": 0.9471575501429165, + "grad_norm": 0.0010571896564215422, + "learning_rate": 0.15725424828340331, + "loss": 0.0864, + "num_input_tokens_seen": 25137216, + "step": 19385 + }, + { + "epoch": 0.9474018518066109, + "grad_norm": 0.0010327696800231934, + "learning_rate": 0.15719541178844715, + "loss": 0.0929, + "num_input_tokens_seen": 25143808, + "step": 19390 + }, + { + "epoch": 0.9476461534703051, + "grad_norm": 0.0013707259204238653, + "learning_rate": 0.15713657418386806, + "loss": 0.1123, + "num_input_tokens_seen": 25150112, + "step": 19395 + }, + { + "epoch": 0.9478904551339995, + "grad_norm": 0.001999193336814642, + "learning_rate": 0.15707773547873957, + "loss": 0.1089, + "num_input_tokens_seen": 25156480, + "step": 19400 + }, + { + "epoch": 0.9478904551339995, + "eval_loss": 0.09777997434139252, + "eval_runtime": 401.9063, + "eval_samples_per_second": 90.531, + "eval_steps_per_second": 22.635, + "num_input_tokens_seen": 25156480, + "step": 19400 + }, + { + "epoch": 0.9481347567976938, + "grad_norm": 0.001270540407858789, + "learning_rate": 0.1570188956821353, + "loss": 0.093, + "num_input_tokens_seen": 25162400, + "step": 19405 + }, + { + "epoch": 0.9483790584613881, + "grad_norm": 0.0008657630532979965, + "learning_rate": 0.1569600548031291, + "loss": 0.0848, + "num_input_tokens_seen": 25168608, + "step": 19410 + }, + { + "epoch": 0.9486233601250824, + "grad_norm": 0.0008915922953747213, + "learning_rate": 0.156901212850795, + "loss": 0.0677, + "num_input_tokens_seen": 25175104, + "step": 19415 + }, + { + "epoch": 0.9488676617887768, + "grad_norm": 0.0016577885253354907, + "learning_rate": 0.15684236983420716, + "loss": 0.1109, + "num_input_tokens_seen": 25181536, + "step": 19420 + }, + { + "epoch": 0.9491119634524711, + "grad_norm": 0.0013089071726426482, + "learning_rate": 0.1567835257624399, + "loss": 0.1004, + "num_input_tokens_seen": 25187808, + "step": 19425 + }, + { + "epoch": 0.9493562651161654, + "grad_norm": 0.0009275113116018474, + "learning_rate": 0.1567246806445677, + "loss": 0.0734, + "num_input_tokens_seen": 25193888, + "step": 19430 + }, + { + "epoch": 0.9496005667798598, + "grad_norm": 0.0023378075566142797, + "learning_rate": 0.15666583448966526, + "loss": 0.1154, + "num_input_tokens_seen": 25200288, + "step": 19435 + }, + { + "epoch": 0.9498448684435541, + "grad_norm": 0.0009703518007881939, + "learning_rate": 0.1566069873068074, + "loss": 0.069, + "num_input_tokens_seen": 25206848, + "step": 19440 + }, + { + "epoch": 0.9500891701072485, + "grad_norm": 0.002438964322209358, + "learning_rate": 0.156548139105069, + "loss": 0.1308, + "num_input_tokens_seen": 25212864, + "step": 19445 + }, + { + "epoch": 0.9503334717709427, + "grad_norm": 0.0008987521287053823, + "learning_rate": 0.15648928989352529, + "loss": 0.1021, + "num_input_tokens_seen": 25219200, + "step": 19450 + }, + { + "epoch": 0.9505777734346371, + "grad_norm": 0.0010104147950187325, + "learning_rate": 0.15643043968125156, + "loss": 0.1117, + "num_input_tokens_seen": 25226240, + "step": 19455 + }, + { + "epoch": 0.9508220750983314, + "grad_norm": 0.000831836718134582, + "learning_rate": 0.15637158847732316, + "loss": 0.095, + "num_input_tokens_seen": 25232576, + "step": 19460 + }, + { + "epoch": 0.9510663767620258, + "grad_norm": 0.001205036765895784, + "learning_rate": 0.15631273629081582, + "loss": 0.1029, + "num_input_tokens_seen": 25239136, + "step": 19465 + }, + { + "epoch": 0.95131067842572, + "grad_norm": 0.0016789153451099992, + "learning_rate": 0.15625388313080518, + "loss": 0.1072, + "num_input_tokens_seen": 25245632, + "step": 19470 + }, + { + "epoch": 0.9515549800894144, + "grad_norm": 0.0008196373819373548, + "learning_rate": 0.15619502900636714, + "loss": 0.0719, + "num_input_tokens_seen": 25252000, + "step": 19475 + }, + { + "epoch": 0.9517992817531087, + "grad_norm": 0.0009039200958795846, + "learning_rate": 0.15613617392657783, + "loss": 0.1016, + "num_input_tokens_seen": 25258080, + "step": 19480 + }, + { + "epoch": 0.9520435834168031, + "grad_norm": 0.0027167373336851597, + "learning_rate": 0.15607731790051335, + "loss": 0.1193, + "num_input_tokens_seen": 25264256, + "step": 19485 + }, + { + "epoch": 0.9522878850804974, + "grad_norm": 0.0012281883973628283, + "learning_rate": 0.15601846093725008, + "loss": 0.099, + "num_input_tokens_seen": 25270496, + "step": 19490 + }, + { + "epoch": 0.9525321867441917, + "grad_norm": 0.0009775864891707897, + "learning_rate": 0.1559596030458645, + "loss": 0.0957, + "num_input_tokens_seen": 25276864, + "step": 19495 + }, + { + "epoch": 0.9527764884078861, + "grad_norm": 0.000571657728869468, + "learning_rate": 0.1559007442354333, + "loss": 0.0968, + "num_input_tokens_seen": 25283520, + "step": 19500 + }, + { + "epoch": 0.9530207900715804, + "grad_norm": 0.0010720273712649941, + "learning_rate": 0.15584188451503314, + "loss": 0.1205, + "num_input_tokens_seen": 25289600, + "step": 19505 + }, + { + "epoch": 0.9532650917352747, + "grad_norm": 0.0029395383317023516, + "learning_rate": 0.15578302389374094, + "loss": 0.116, + "num_input_tokens_seen": 25296416, + "step": 19510 + }, + { + "epoch": 0.953509393398969, + "grad_norm": 0.0010829437524080276, + "learning_rate": 0.1557241623806338, + "loss": 0.0892, + "num_input_tokens_seen": 25303136, + "step": 19515 + }, + { + "epoch": 0.9537536950626634, + "grad_norm": 0.0010911535937339067, + "learning_rate": 0.15566529998478887, + "loss": 0.1021, + "num_input_tokens_seen": 25309920, + "step": 19520 + }, + { + "epoch": 0.9539979967263577, + "grad_norm": 0.0010262068826705217, + "learning_rate": 0.15560643671528354, + "loss": 0.0878, + "num_input_tokens_seen": 25316768, + "step": 19525 + }, + { + "epoch": 0.954242298390052, + "grad_norm": 0.000418227311456576, + "learning_rate": 0.15554757258119514, + "loss": 0.081, + "num_input_tokens_seen": 25323552, + "step": 19530 + }, + { + "epoch": 0.9544866000537464, + "grad_norm": 0.0010059904307126999, + "learning_rate": 0.1554887075916014, + "loss": 0.0813, + "num_input_tokens_seen": 25330112, + "step": 19535 + }, + { + "epoch": 0.9547309017174407, + "grad_norm": 0.000989248394034803, + "learning_rate": 0.15542984175558, + "loss": 0.0964, + "num_input_tokens_seen": 25336416, + "step": 19540 + }, + { + "epoch": 0.9549752033811351, + "grad_norm": 0.0009396843961440027, + "learning_rate": 0.1553709750822087, + "loss": 0.0963, + "num_input_tokens_seen": 25342880, + "step": 19545 + }, + { + "epoch": 0.9552195050448293, + "grad_norm": 0.0009968355298042297, + "learning_rate": 0.15531210758056554, + "loss": 0.0955, + "num_input_tokens_seen": 25349408, + "step": 19550 + }, + { + "epoch": 0.9554638067085237, + "grad_norm": 0.0010713570518419147, + "learning_rate": 0.15525323925972867, + "loss": 0.0679, + "num_input_tokens_seen": 25355744, + "step": 19555 + }, + { + "epoch": 0.955708108372218, + "grad_norm": 0.0016800948651507497, + "learning_rate": 0.15519437012877627, + "loss": 0.0963, + "num_input_tokens_seen": 25362496, + "step": 19560 + }, + { + "epoch": 0.9559524100359124, + "grad_norm": 0.0009568536188453436, + "learning_rate": 0.15513550019678676, + "loss": 0.0933, + "num_input_tokens_seen": 25369088, + "step": 19565 + }, + { + "epoch": 0.9561967116996066, + "grad_norm": 0.0015514919068664312, + "learning_rate": 0.15507662947283854, + "loss": 0.1008, + "num_input_tokens_seen": 25376160, + "step": 19570 + }, + { + "epoch": 0.956441013363301, + "grad_norm": 0.0009099450544454157, + "learning_rate": 0.15501775796601028, + "loss": 0.0778, + "num_input_tokens_seen": 25382528, + "step": 19575 + }, + { + "epoch": 0.9566853150269954, + "grad_norm": 0.00046951722470112145, + "learning_rate": 0.15495888568538066, + "loss": 0.0955, + "num_input_tokens_seen": 25389632, + "step": 19580 + }, + { + "epoch": 0.9569296166906897, + "grad_norm": 0.0013182936236262321, + "learning_rate": 0.1549000126400286, + "loss": 0.0895, + "num_input_tokens_seen": 25396608, + "step": 19585 + }, + { + "epoch": 0.957173918354384, + "grad_norm": 0.0017430984880775213, + "learning_rate": 0.15484113883903294, + "loss": 0.1247, + "num_input_tokens_seen": 25403264, + "step": 19590 + }, + { + "epoch": 0.9574182200180783, + "grad_norm": 0.0011052499758079648, + "learning_rate": 0.15478226429147288, + "loss": 0.0992, + "num_input_tokens_seen": 25409824, + "step": 19595 + }, + { + "epoch": 0.9576625216817727, + "grad_norm": 0.0013589917216449976, + "learning_rate": 0.15472338900642757, + "loss": 0.1263, + "num_input_tokens_seen": 25415936, + "step": 19600 + }, + { + "epoch": 0.9576625216817727, + "eval_loss": 0.09705813974142075, + "eval_runtime": 402.0489, + "eval_samples_per_second": 90.499, + "eval_steps_per_second": 22.627, + "num_input_tokens_seen": 25415936, + "step": 19600 + }, + { + "epoch": 0.957906823345467, + "grad_norm": 0.0010052158031612635, + "learning_rate": 0.15466451299297632, + "loss": 0.0911, + "num_input_tokens_seen": 25422240, + "step": 19605 + }, + { + "epoch": 0.9581511250091613, + "grad_norm": 0.0012780413962900639, + "learning_rate": 0.15460563626019852, + "loss": 0.1151, + "num_input_tokens_seen": 25428992, + "step": 19610 + }, + { + "epoch": 0.9583954266728556, + "grad_norm": 0.0011128669138997793, + "learning_rate": 0.15454675881717375, + "loss": 0.0819, + "num_input_tokens_seen": 25435488, + "step": 19615 + }, + { + "epoch": 0.95863972833655, + "grad_norm": 0.0010166842257604003, + "learning_rate": 0.1544878806729816, + "loss": 0.0929, + "num_input_tokens_seen": 25441664, + "step": 19620 + }, + { + "epoch": 0.9588840300002442, + "grad_norm": 0.0009239534265361726, + "learning_rate": 0.1544290018367019, + "loss": 0.0908, + "num_input_tokens_seen": 25448608, + "step": 19625 + }, + { + "epoch": 0.9591283316639386, + "grad_norm": 0.0005114906816743314, + "learning_rate": 0.15437012231741445, + "loss": 0.0937, + "num_input_tokens_seen": 25455232, + "step": 19630 + }, + { + "epoch": 0.959372633327633, + "grad_norm": 0.0009685500990599394, + "learning_rate": 0.1543112421241992, + "loss": 0.0657, + "num_input_tokens_seen": 25461440, + "step": 19635 + }, + { + "epoch": 0.9596169349913273, + "grad_norm": 0.0006362772546708584, + "learning_rate": 0.15425236126613626, + "loss": 0.0949, + "num_input_tokens_seen": 25467776, + "step": 19640 + }, + { + "epoch": 0.9598612366550217, + "grad_norm": 0.0006737885996699333, + "learning_rate": 0.15419347975230577, + "loss": 0.1057, + "num_input_tokens_seen": 25474080, + "step": 19645 + }, + { + "epoch": 0.9601055383187159, + "grad_norm": 0.0016124688554555178, + "learning_rate": 0.154134597591788, + "loss": 0.0954, + "num_input_tokens_seen": 25480768, + "step": 19650 + }, + { + "epoch": 0.9603498399824103, + "grad_norm": 0.0015156905865296721, + "learning_rate": 0.1540757147936633, + "loss": 0.0858, + "num_input_tokens_seen": 25487520, + "step": 19655 + }, + { + "epoch": 0.9605941416461046, + "grad_norm": 0.0012522597098723054, + "learning_rate": 0.1540168313670122, + "loss": 0.0756, + "num_input_tokens_seen": 25494016, + "step": 19660 + }, + { + "epoch": 0.960838443309799, + "grad_norm": 0.0011643068864941597, + "learning_rate": 0.1539579473209152, + "loss": 0.0839, + "num_input_tokens_seen": 25500800, + "step": 19665 + }, + { + "epoch": 0.9610827449734932, + "grad_norm": 0.0010584724368527532, + "learning_rate": 0.15389906266445294, + "loss": 0.0909, + "num_input_tokens_seen": 25506848, + "step": 19670 + }, + { + "epoch": 0.9613270466371876, + "grad_norm": 0.0007148032891564071, + "learning_rate": 0.15384017740670627, + "loss": 0.0857, + "num_input_tokens_seen": 25513184, + "step": 19675 + }, + { + "epoch": 0.961571348300882, + "grad_norm": 0.002132401568815112, + "learning_rate": 0.15378129155675602, + "loss": 0.1298, + "num_input_tokens_seen": 25519552, + "step": 19680 + }, + { + "epoch": 0.9618156499645762, + "grad_norm": 0.0011281768092885613, + "learning_rate": 0.15372240512368307, + "loss": 0.1197, + "num_input_tokens_seen": 25525632, + "step": 19685 + }, + { + "epoch": 0.9620599516282706, + "grad_norm": 0.0016916365129873157, + "learning_rate": 0.1536635181165684, + "loss": 0.0986, + "num_input_tokens_seen": 25532032, + "step": 19690 + }, + { + "epoch": 0.9623042532919649, + "grad_norm": 0.0021551765967160463, + "learning_rate": 0.15360463054449328, + "loss": 0.101, + "num_input_tokens_seen": 25538272, + "step": 19695 + }, + { + "epoch": 0.9625485549556593, + "grad_norm": 0.001363311312161386, + "learning_rate": 0.1535457424165388, + "loss": 0.1094, + "num_input_tokens_seen": 25544640, + "step": 19700 + }, + { + "epoch": 0.9627928566193535, + "grad_norm": 0.001247812993824482, + "learning_rate": 0.15348685374178628, + "loss": 0.1161, + "num_input_tokens_seen": 25551456, + "step": 19705 + }, + { + "epoch": 0.9630371582830479, + "grad_norm": 0.0011072542984038591, + "learning_rate": 0.1534279645293171, + "loss": 0.0936, + "num_input_tokens_seen": 25558016, + "step": 19710 + }, + { + "epoch": 0.9632814599467422, + "grad_norm": 0.0009158953325822949, + "learning_rate": 0.1533690747882127, + "loss": 0.0836, + "num_input_tokens_seen": 25564544, + "step": 19715 + }, + { + "epoch": 0.9635257616104366, + "grad_norm": 0.0008117787074297667, + "learning_rate": 0.15331018452755465, + "loss": 0.0915, + "num_input_tokens_seen": 25570912, + "step": 19720 + }, + { + "epoch": 0.963770063274131, + "grad_norm": 0.0017150136409327388, + "learning_rate": 0.15325129375642457, + "loss": 0.1267, + "num_input_tokens_seen": 25577632, + "step": 19725 + }, + { + "epoch": 0.9640143649378252, + "grad_norm": 0.0024172391276806593, + "learning_rate": 0.15319240248390406, + "loss": 0.1417, + "num_input_tokens_seen": 25583936, + "step": 19730 + }, + { + "epoch": 0.9642586666015196, + "grad_norm": 0.001384394010528922, + "learning_rate": 0.153133510719075, + "loss": 0.1115, + "num_input_tokens_seen": 25590656, + "step": 19735 + }, + { + "epoch": 0.9645029682652139, + "grad_norm": 0.0011881644604727626, + "learning_rate": 0.15307461847101922, + "loss": 0.1087, + "num_input_tokens_seen": 25596864, + "step": 19740 + }, + { + "epoch": 0.9647472699289082, + "grad_norm": 0.0008007743745110929, + "learning_rate": 0.15301572574881864, + "loss": 0.0839, + "num_input_tokens_seen": 25603328, + "step": 19745 + }, + { + "epoch": 0.9649915715926025, + "grad_norm": 0.0009673970052972436, + "learning_rate": 0.15295683256155523, + "loss": 0.0968, + "num_input_tokens_seen": 25610272, + "step": 19750 + }, + { + "epoch": 0.9652358732562969, + "grad_norm": 0.0009568413370288908, + "learning_rate": 0.15289793891831113, + "loss": 0.1096, + "num_input_tokens_seen": 25616896, + "step": 19755 + }, + { + "epoch": 0.9654801749199912, + "grad_norm": 0.0013360594166442752, + "learning_rate": 0.15283904482816837, + "loss": 0.1059, + "num_input_tokens_seen": 25623456, + "step": 19760 + }, + { + "epoch": 0.9657244765836855, + "grad_norm": 0.0007826304063200951, + "learning_rate": 0.15278015030020928, + "loss": 0.0964, + "num_input_tokens_seen": 25629504, + "step": 19765 + }, + { + "epoch": 0.9659687782473798, + "grad_norm": 0.0009999366011470556, + "learning_rate": 0.152721255343516, + "loss": 0.1173, + "num_input_tokens_seen": 25635776, + "step": 19770 + }, + { + "epoch": 0.9662130799110742, + "grad_norm": 0.0010328178759664297, + "learning_rate": 0.15266235996717098, + "loss": 0.0969, + "num_input_tokens_seen": 25642272, + "step": 19775 + }, + { + "epoch": 0.9664573815747686, + "grad_norm": 0.000902714382391423, + "learning_rate": 0.15260346418025664, + "loss": 0.0889, + "num_input_tokens_seen": 25648864, + "step": 19780 + }, + { + "epoch": 0.9667016832384628, + "grad_norm": 0.00084716483252123, + "learning_rate": 0.15254456799185537, + "loss": 0.082, + "num_input_tokens_seen": 25655968, + "step": 19785 + }, + { + "epoch": 0.9669459849021572, + "grad_norm": 0.0011574591044336557, + "learning_rate": 0.15248567141104974, + "loss": 0.1147, + "num_input_tokens_seen": 25663456, + "step": 19790 + }, + { + "epoch": 0.9671902865658515, + "grad_norm": 0.0012633054284378886, + "learning_rate": 0.15242677444692232, + "loss": 0.1102, + "num_input_tokens_seen": 25670464, + "step": 19795 + }, + { + "epoch": 0.9674345882295459, + "grad_norm": 0.0009620675700716674, + "learning_rate": 0.15236787710855584, + "loss": 0.1038, + "num_input_tokens_seen": 25677472, + "step": 19800 + }, + { + "epoch": 0.9674345882295459, + "eval_loss": 0.0984303206205368, + "eval_runtime": 402.346, + "eval_samples_per_second": 90.432, + "eval_steps_per_second": 22.61, + "num_input_tokens_seen": 25677472, + "step": 19800 + }, + { + "epoch": 0.9676788898932401, + "grad_norm": 0.0008570508216507733, + "learning_rate": 0.1523089794050329, + "loss": 0.0954, + "num_input_tokens_seen": 25683584, + "step": 19805 + }, + { + "epoch": 0.9679231915569345, + "grad_norm": 0.0011416887864470482, + "learning_rate": 0.15225008134543633, + "loss": 0.1162, + "num_input_tokens_seen": 25689984, + "step": 19810 + }, + { + "epoch": 0.9681674932206288, + "grad_norm": 0.0012827160535380244, + "learning_rate": 0.15219118293884895, + "loss": 0.09, + "num_input_tokens_seen": 25696608, + "step": 19815 + }, + { + "epoch": 0.9684117948843232, + "grad_norm": 0.0009720964590087533, + "learning_rate": 0.15213228419435362, + "loss": 0.0959, + "num_input_tokens_seen": 25703488, + "step": 19820 + }, + { + "epoch": 0.9686560965480175, + "grad_norm": 0.0009992228588089347, + "learning_rate": 0.15207338512103327, + "loss": 0.0972, + "num_input_tokens_seen": 25709728, + "step": 19825 + }, + { + "epoch": 0.9689003982117118, + "grad_norm": 0.0014242333127185702, + "learning_rate": 0.1520144857279709, + "loss": 0.0859, + "num_input_tokens_seen": 25716320, + "step": 19830 + }, + { + "epoch": 0.9691446998754062, + "grad_norm": 0.001232158625498414, + "learning_rate": 0.1519555860242495, + "loss": 0.0984, + "num_input_tokens_seen": 25722528, + "step": 19835 + }, + { + "epoch": 0.9693890015391005, + "grad_norm": 0.00073143100598827, + "learning_rate": 0.15189668601895218, + "loss": 0.0682, + "num_input_tokens_seen": 25728864, + "step": 19840 + }, + { + "epoch": 0.9696333032027948, + "grad_norm": 0.0018125054193660617, + "learning_rate": 0.151837785721162, + "loss": 0.1116, + "num_input_tokens_seen": 25735264, + "step": 19845 + }, + { + "epoch": 0.9698776048664891, + "grad_norm": 0.0012785980943590403, + "learning_rate": 0.15177888513996218, + "loss": 0.0809, + "num_input_tokens_seen": 25741536, + "step": 19850 + }, + { + "epoch": 0.9701219065301835, + "grad_norm": 0.0010037146275863051, + "learning_rate": 0.15171998428443592, + "loss": 0.0831, + "num_input_tokens_seen": 25748000, + "step": 19855 + }, + { + "epoch": 0.9703662081938778, + "grad_norm": 0.0016373117687180638, + "learning_rate": 0.1516610831636665, + "loss": 0.0997, + "num_input_tokens_seen": 25754304, + "step": 19860 + }, + { + "epoch": 0.9706105098575721, + "grad_norm": 0.0013300738064572215, + "learning_rate": 0.15160218178673715, + "loss": 0.126, + "num_input_tokens_seen": 25760640, + "step": 19865 + }, + { + "epoch": 0.9708548115212665, + "grad_norm": 0.0016958208288997412, + "learning_rate": 0.15154328016273122, + "loss": 0.0832, + "num_input_tokens_seen": 25767136, + "step": 19870 + }, + { + "epoch": 0.9710991131849608, + "grad_norm": 0.0009533455013297498, + "learning_rate": 0.1514843783007321, + "loss": 0.0857, + "num_input_tokens_seen": 25773984, + "step": 19875 + }, + { + "epoch": 0.9713434148486552, + "grad_norm": 0.0010339895961806178, + "learning_rate": 0.15142547620982322, + "loss": 0.1439, + "num_input_tokens_seen": 25780128, + "step": 19880 + }, + { + "epoch": 0.9715877165123494, + "grad_norm": 0.0012152347480878234, + "learning_rate": 0.15136657389908797, + "loss": 0.0914, + "num_input_tokens_seen": 25786336, + "step": 19885 + }, + { + "epoch": 0.9718320181760438, + "grad_norm": 0.0011286747176200151, + "learning_rate": 0.15130767137760986, + "loss": 0.1122, + "num_input_tokens_seen": 25792704, + "step": 19890 + }, + { + "epoch": 0.9720763198397381, + "grad_norm": 0.0009636828908696771, + "learning_rate": 0.15124876865447243, + "loss": 0.0973, + "num_input_tokens_seen": 25799136, + "step": 19895 + }, + { + "epoch": 0.9723206215034325, + "grad_norm": 0.0011533248471096158, + "learning_rate": 0.15118986573875912, + "loss": 0.0912, + "num_input_tokens_seen": 25805888, + "step": 19900 + }, + { + "epoch": 0.9725649231671267, + "grad_norm": 0.0008717195596545935, + "learning_rate": 0.15113096263955358, + "loss": 0.1038, + "num_input_tokens_seen": 25812000, + "step": 19905 + }, + { + "epoch": 0.9728092248308211, + "grad_norm": 0.0006454290705733001, + "learning_rate": 0.1510720593659394, + "loss": 0.0955, + "num_input_tokens_seen": 25818880, + "step": 19910 + }, + { + "epoch": 0.9730535264945154, + "grad_norm": 0.0007064878009259701, + "learning_rate": 0.15101315592700015, + "loss": 0.0828, + "num_input_tokens_seen": 25825600, + "step": 19915 + }, + { + "epoch": 0.9732978281582098, + "grad_norm": 0.0009175879531539977, + "learning_rate": 0.15095425233181956, + "loss": 0.0986, + "num_input_tokens_seen": 25831840, + "step": 19920 + }, + { + "epoch": 0.9735421298219041, + "grad_norm": 0.0010531549341976643, + "learning_rate": 0.15089534858948128, + "loss": 0.1045, + "num_input_tokens_seen": 25838528, + "step": 19925 + }, + { + "epoch": 0.9737864314855984, + "grad_norm": 0.0007368625374510884, + "learning_rate": 0.15083644470906898, + "loss": 0.0797, + "num_input_tokens_seen": 25845120, + "step": 19930 + }, + { + "epoch": 0.9740307331492928, + "grad_norm": 0.0009173737489618361, + "learning_rate": 0.1507775406996664, + "loss": 0.1074, + "num_input_tokens_seen": 25851424, + "step": 19935 + }, + { + "epoch": 0.974275034812987, + "grad_norm": 0.0007408498204313219, + "learning_rate": 0.15071863657035725, + "loss": 0.0959, + "num_input_tokens_seen": 25858176, + "step": 19940 + }, + { + "epoch": 0.9745193364766814, + "grad_norm": 0.0010209650499746203, + "learning_rate": 0.15065973233022534, + "loss": 0.1054, + "num_input_tokens_seen": 25864672, + "step": 19945 + }, + { + "epoch": 0.9747636381403757, + "grad_norm": 0.0015023340238258243, + "learning_rate": 0.15060082798835442, + "loss": 0.1001, + "num_input_tokens_seen": 25871136, + "step": 19950 + }, + { + "epoch": 0.9750079398040701, + "grad_norm": 0.0008068465976975858, + "learning_rate": 0.15054192355382823, + "loss": 0.0748, + "num_input_tokens_seen": 25877664, + "step": 19955 + }, + { + "epoch": 0.9752522414677643, + "grad_norm": 0.0010661095147952437, + "learning_rate": 0.15048301903573066, + "loss": 0.0981, + "num_input_tokens_seen": 25884256, + "step": 19960 + }, + { + "epoch": 0.9754965431314587, + "grad_norm": 0.001136621693149209, + "learning_rate": 0.15042411444314546, + "loss": 0.0836, + "num_input_tokens_seen": 25890464, + "step": 19965 + }, + { + "epoch": 0.9757408447951531, + "grad_norm": 0.000739955110475421, + "learning_rate": 0.1503652097851565, + "loss": 0.0815, + "num_input_tokens_seen": 25896800, + "step": 19970 + }, + { + "epoch": 0.9759851464588474, + "grad_norm": 0.0009603677317500114, + "learning_rate": 0.15030630507084758, + "loss": 0.0991, + "num_input_tokens_seen": 25902752, + "step": 19975 + }, + { + "epoch": 0.9762294481225418, + "grad_norm": 0.0011389831779524684, + "learning_rate": 0.1502474003093026, + "loss": 0.1171, + "num_input_tokens_seen": 25908992, + "step": 19980 + }, + { + "epoch": 0.976473749786236, + "grad_norm": 0.0009523942717351019, + "learning_rate": 0.15018849550960536, + "loss": 0.0798, + "num_input_tokens_seen": 25915392, + "step": 19985 + }, + { + "epoch": 0.9767180514499304, + "grad_norm": 0.0012472537346184254, + "learning_rate": 0.15012959068083975, + "loss": 0.0884, + "num_input_tokens_seen": 25921632, + "step": 19990 + }, + { + "epoch": 0.9769623531136247, + "grad_norm": 0.0013965652324259281, + "learning_rate": 0.1500706858320896, + "loss": 0.1021, + "num_input_tokens_seen": 25928032, + "step": 19995 + }, + { + "epoch": 0.977206654777319, + "grad_norm": 0.0008893098565749824, + "learning_rate": 0.15001178097243886, + "loss": 0.0826, + "num_input_tokens_seen": 25934656, + "step": 20000 + }, + { + "epoch": 0.977206654777319, + "eval_loss": 0.09820657968521118, + "eval_runtime": 402.4796, + "eval_samples_per_second": 90.402, + "eval_steps_per_second": 22.602, + "num_input_tokens_seen": 25934656, + "step": 20000 + }, + { + "epoch": 0.9774509564410133, + "grad_norm": 0.0016499778721481562, + "learning_rate": 0.1499528761109713, + "loss": 0.1225, + "num_input_tokens_seen": 25941312, + "step": 20005 + }, + { + "epoch": 0.9776952581047077, + "grad_norm": 0.0015580557519569993, + "learning_rate": 0.14989397125677087, + "loss": 0.1116, + "num_input_tokens_seen": 25947520, + "step": 20010 + }, + { + "epoch": 0.9779395597684021, + "grad_norm": 0.0015733519103378057, + "learning_rate": 0.14983506641892141, + "loss": 0.0876, + "num_input_tokens_seen": 25953856, + "step": 20015 + }, + { + "epoch": 0.9781838614320963, + "grad_norm": 0.001107610878534615, + "learning_rate": 0.14977616160650672, + "loss": 0.1074, + "num_input_tokens_seen": 25961120, + "step": 20020 + }, + { + "epoch": 0.9784281630957907, + "grad_norm": 0.0007070951396599412, + "learning_rate": 0.14971725682861076, + "loss": 0.0903, + "num_input_tokens_seen": 25967584, + "step": 20025 + }, + { + "epoch": 0.978672464759485, + "grad_norm": 0.0017815284663811326, + "learning_rate": 0.14965835209431738, + "loss": 0.0769, + "num_input_tokens_seen": 25973824, + "step": 20030 + }, + { + "epoch": 0.9789167664231794, + "grad_norm": 0.001036068657413125, + "learning_rate": 0.14959944741271036, + "loss": 0.1092, + "num_input_tokens_seen": 25980288, + "step": 20035 + }, + { + "epoch": 0.9791610680868736, + "grad_norm": 0.001214879914186895, + "learning_rate": 0.14954054279287363, + "loss": 0.0831, + "num_input_tokens_seen": 25986720, + "step": 20040 + }, + { + "epoch": 0.979405369750568, + "grad_norm": 0.0014828440034762025, + "learning_rate": 0.14948163824389094, + "loss": 0.0978, + "num_input_tokens_seen": 25993216, + "step": 20045 + }, + { + "epoch": 0.9796496714142623, + "grad_norm": 0.0009042855235747993, + "learning_rate": 0.14942273377484613, + "loss": 0.089, + "num_input_tokens_seen": 25999552, + "step": 20050 + }, + { + "epoch": 0.9798939730779567, + "grad_norm": 0.0012420243583619595, + "learning_rate": 0.1493638293948231, + "loss": 0.0807, + "num_input_tokens_seen": 26005664, + "step": 20055 + }, + { + "epoch": 0.9801382747416509, + "grad_norm": 0.0011997104156762362, + "learning_rate": 0.14930492511290547, + "loss": 0.1165, + "num_input_tokens_seen": 26012704, + "step": 20060 + }, + { + "epoch": 0.9803825764053453, + "grad_norm": 0.0007684106822125614, + "learning_rate": 0.14924602093817715, + "loss": 0.1017, + "num_input_tokens_seen": 26019488, + "step": 20065 + }, + { + "epoch": 0.9806268780690397, + "grad_norm": 0.0010029751574620605, + "learning_rate": 0.14918711687972194, + "loss": 0.1146, + "num_input_tokens_seen": 26025792, + "step": 20070 + }, + { + "epoch": 0.980871179732734, + "grad_norm": 0.0011154826497659087, + "learning_rate": 0.14912821294662346, + "loss": 0.1078, + "num_input_tokens_seen": 26031648, + "step": 20075 + }, + { + "epoch": 0.9811154813964283, + "grad_norm": 0.0007222027634270489, + "learning_rate": 0.14906930914796554, + "loss": 0.073, + "num_input_tokens_seen": 26038048, + "step": 20080 + }, + { + "epoch": 0.9813597830601226, + "grad_norm": 0.0013214248465374112, + "learning_rate": 0.14901040549283182, + "loss": 0.1162, + "num_input_tokens_seen": 26044608, + "step": 20085 + }, + { + "epoch": 0.981604084723817, + "grad_norm": 0.0015496048144996166, + "learning_rate": 0.148951501990306, + "loss": 0.0901, + "num_input_tokens_seen": 26051264, + "step": 20090 + }, + { + "epoch": 0.9818483863875113, + "grad_norm": 0.0008074213401414454, + "learning_rate": 0.14889259864947177, + "loss": 0.0873, + "num_input_tokens_seen": 26058112, + "step": 20095 + }, + { + "epoch": 0.9820926880512056, + "grad_norm": 0.0009300995152443647, + "learning_rate": 0.14883369547941272, + "loss": 0.0732, + "num_input_tokens_seen": 26064160, + "step": 20100 + }, + { + "epoch": 0.9823369897148999, + "grad_norm": 0.0013062324142083526, + "learning_rate": 0.14877479248921247, + "loss": 0.0817, + "num_input_tokens_seen": 26070272, + "step": 20105 + }, + { + "epoch": 0.9825812913785943, + "grad_norm": 0.001362847862765193, + "learning_rate": 0.14871588968795468, + "loss": 0.1011, + "num_input_tokens_seen": 26077120, + "step": 20110 + }, + { + "epoch": 0.9828255930422887, + "grad_norm": 0.0009625280508771539, + "learning_rate": 0.1486569870847228, + "loss": 0.085, + "num_input_tokens_seen": 26083776, + "step": 20115 + }, + { + "epoch": 0.9830698947059829, + "grad_norm": 0.0008829927537590265, + "learning_rate": 0.1485980846886004, + "loss": 0.1039, + "num_input_tokens_seen": 26090016, + "step": 20120 + }, + { + "epoch": 0.9833141963696773, + "grad_norm": 0.0012838944094255567, + "learning_rate": 0.14853918250867096, + "loss": 0.1006, + "num_input_tokens_seen": 26096608, + "step": 20125 + }, + { + "epoch": 0.9835584980333716, + "grad_norm": 0.0008783689700067043, + "learning_rate": 0.1484802805540179, + "loss": 0.0922, + "num_input_tokens_seen": 26102688, + "step": 20130 + }, + { + "epoch": 0.983802799697066, + "grad_norm": 0.000715558126103133, + "learning_rate": 0.14842137883372472, + "loss": 0.0765, + "num_input_tokens_seen": 26108704, + "step": 20135 + }, + { + "epoch": 0.9840471013607602, + "grad_norm": 0.0013695212546736002, + "learning_rate": 0.14836247735687474, + "loss": 0.1077, + "num_input_tokens_seen": 26115360, + "step": 20140 + }, + { + "epoch": 0.9842914030244546, + "grad_norm": 0.0013420708710327744, + "learning_rate": 0.14830357613255132, + "loss": 0.0855, + "num_input_tokens_seen": 26121728, + "step": 20145 + }, + { + "epoch": 0.9845357046881489, + "grad_norm": 0.0007720611174590886, + "learning_rate": 0.1482446751698378, + "loss": 0.0938, + "num_input_tokens_seen": 26128256, + "step": 20150 + }, + { + "epoch": 0.9847800063518433, + "grad_norm": 0.001142480643466115, + "learning_rate": 0.14818577447781744, + "loss": 0.0955, + "num_input_tokens_seen": 26135200, + "step": 20155 + }, + { + "epoch": 0.9850243080155375, + "grad_norm": 0.0010735546238720417, + "learning_rate": 0.14812687406557346, + "loss": 0.0847, + "num_input_tokens_seen": 26141472, + "step": 20160 + }, + { + "epoch": 0.9852686096792319, + "grad_norm": 0.0007197023369371891, + "learning_rate": 0.14806797394218899, + "loss": 0.0882, + "num_input_tokens_seen": 26148064, + "step": 20165 + }, + { + "epoch": 0.9855129113429263, + "grad_norm": 0.0016247391467913985, + "learning_rate": 0.1480090741167472, + "loss": 0.1178, + "num_input_tokens_seen": 26154400, + "step": 20170 + }, + { + "epoch": 0.9857572130066206, + "grad_norm": 0.0009538683807477355, + "learning_rate": 0.1479501745983313, + "loss": 0.0806, + "num_input_tokens_seen": 26160640, + "step": 20175 + }, + { + "epoch": 0.9860015146703149, + "grad_norm": 0.0005861669778823853, + "learning_rate": 0.14789127539602415, + "loss": 0.048, + "num_input_tokens_seen": 26167456, + "step": 20180 + }, + { + "epoch": 0.9862458163340092, + "grad_norm": 0.0019942333456128836, + "learning_rate": 0.14783237651890885, + "loss": 0.1145, + "num_input_tokens_seen": 26174112, + "step": 20185 + }, + { + "epoch": 0.9864901179977036, + "grad_norm": 0.001267713843844831, + "learning_rate": 0.14777347797606838, + "loss": 0.0862, + "num_input_tokens_seen": 26180320, + "step": 20190 + }, + { + "epoch": 0.9867344196613979, + "grad_norm": 0.0016123257810249925, + "learning_rate": 0.14771457977658553, + "loss": 0.0945, + "num_input_tokens_seen": 26186464, + "step": 20195 + }, + { + "epoch": 0.9869787213250922, + "grad_norm": 0.001350108184851706, + "learning_rate": 0.14765568192954326, + "loss": 0.1066, + "num_input_tokens_seen": 26193248, + "step": 20200 + }, + { + "epoch": 0.9869787213250922, + "eval_loss": 0.09710865467786789, + "eval_runtime": 402.299, + "eval_samples_per_second": 90.443, + "eval_steps_per_second": 22.613, + "num_input_tokens_seen": 26193248, + "step": 20200 + }, + { + "epoch": 0.9872230229887865, + "grad_norm": 0.001902968855574727, + "learning_rate": 0.14759678444402421, + "loss": 0.0857, + "num_input_tokens_seen": 26199136, + "step": 20205 + }, + { + "epoch": 0.9874673246524809, + "grad_norm": 0.0007208503084257245, + "learning_rate": 0.14753788732911122, + "loss": 0.0734, + "num_input_tokens_seen": 26205376, + "step": 20210 + }, + { + "epoch": 0.9877116263161753, + "grad_norm": 0.0011088695609942079, + "learning_rate": 0.147478990593887, + "loss": 0.0764, + "num_input_tokens_seen": 26212064, + "step": 20215 + }, + { + "epoch": 0.9879559279798695, + "grad_norm": 0.0020893665496259928, + "learning_rate": 0.14742009424743405, + "loss": 0.1135, + "num_input_tokens_seen": 26218144, + "step": 20220 + }, + { + "epoch": 0.9882002296435639, + "grad_norm": 0.0011056206421926618, + "learning_rate": 0.14736119829883504, + "loss": 0.091, + "num_input_tokens_seen": 26224672, + "step": 20225 + }, + { + "epoch": 0.9884445313072582, + "grad_norm": 0.0008710320107638836, + "learning_rate": 0.14730230275717243, + "loss": 0.1014, + "num_input_tokens_seen": 26230688, + "step": 20230 + }, + { + "epoch": 0.9886888329709526, + "grad_norm": 0.001469020382501185, + "learning_rate": 0.14724340763152854, + "loss": 0.1119, + "num_input_tokens_seen": 26236736, + "step": 20235 + }, + { + "epoch": 0.9889331346346468, + "grad_norm": 0.0006936913705430925, + "learning_rate": 0.14718451293098594, + "loss": 0.097, + "num_input_tokens_seen": 26242688, + "step": 20240 + }, + { + "epoch": 0.9891774362983412, + "grad_norm": 0.001370286918245256, + "learning_rate": 0.14712561866462676, + "loss": 0.1125, + "num_input_tokens_seen": 26249600, + "step": 20245 + }, + { + "epoch": 0.9894217379620355, + "grad_norm": 0.0009414865635335445, + "learning_rate": 0.1470667248415333, + "loss": 0.0844, + "num_input_tokens_seen": 26256032, + "step": 20250 + }, + { + "epoch": 0.9896660396257299, + "grad_norm": 0.0009352503111585975, + "learning_rate": 0.1470078314707878, + "loss": 0.0991, + "num_input_tokens_seen": 26262272, + "step": 20255 + }, + { + "epoch": 0.9899103412894242, + "grad_norm": 0.0012254606699571013, + "learning_rate": 0.14694893856147223, + "loss": 0.0918, + "num_input_tokens_seen": 26269120, + "step": 20260 + }, + { + "epoch": 0.9901546429531185, + "grad_norm": 0.000787491851951927, + "learning_rate": 0.14689004612266868, + "loss": 0.0888, + "num_input_tokens_seen": 26275776, + "step": 20265 + }, + { + "epoch": 0.9903989446168129, + "grad_norm": 0.0007484601810574532, + "learning_rate": 0.14683115416345913, + "loss": 0.1174, + "num_input_tokens_seen": 26282272, + "step": 20270 + }, + { + "epoch": 0.9906432462805072, + "grad_norm": 0.0010564613621681929, + "learning_rate": 0.1467722626929254, + "loss": 0.1147, + "num_input_tokens_seen": 26288416, + "step": 20275 + }, + { + "epoch": 0.9908875479442015, + "grad_norm": 0.0010194072965532541, + "learning_rate": 0.14671337172014937, + "loss": 0.0881, + "num_input_tokens_seen": 26294240, + "step": 20280 + }, + { + "epoch": 0.9911318496078958, + "grad_norm": 0.0008801919175311923, + "learning_rate": 0.14665448125421265, + "loss": 0.0912, + "num_input_tokens_seen": 26300288, + "step": 20285 + }, + { + "epoch": 0.9913761512715902, + "grad_norm": 0.0010182007681578398, + "learning_rate": 0.146595591304197, + "loss": 0.0977, + "num_input_tokens_seen": 26306624, + "step": 20290 + }, + { + "epoch": 0.9916204529352844, + "grad_norm": 0.0007369139930233359, + "learning_rate": 0.14653670187918397, + "loss": 0.1049, + "num_input_tokens_seen": 26312832, + "step": 20295 + }, + { + "epoch": 0.9918647545989788, + "grad_norm": 0.001112372032366693, + "learning_rate": 0.14647781298825502, + "loss": 0.1174, + "num_input_tokens_seen": 26318752, + "step": 20300 + }, + { + "epoch": 0.9921090562626731, + "grad_norm": 0.001292865606956184, + "learning_rate": 0.14641892464049153, + "loss": 0.0952, + "num_input_tokens_seen": 26325120, + "step": 20305 + }, + { + "epoch": 0.9923533579263675, + "grad_norm": 0.0009079300216399133, + "learning_rate": 0.14636003684497495, + "loss": 0.087, + "num_input_tokens_seen": 26331744, + "step": 20310 + }, + { + "epoch": 0.9925976595900619, + "grad_norm": 0.0009284615516662598, + "learning_rate": 0.14630114961078636, + "loss": 0.1013, + "num_input_tokens_seen": 26338112, + "step": 20315 + }, + { + "epoch": 0.9928419612537561, + "grad_norm": 0.0018358345841988921, + "learning_rate": 0.14624226294700704, + "loss": 0.0871, + "num_input_tokens_seen": 26344800, + "step": 20320 + }, + { + "epoch": 0.9930862629174505, + "grad_norm": 0.0011026861611753702, + "learning_rate": 0.14618337686271793, + "loss": 0.0937, + "num_input_tokens_seen": 26351200, + "step": 20325 + }, + { + "epoch": 0.9933305645811448, + "grad_norm": 0.0010634901700541377, + "learning_rate": 0.1461244913670001, + "loss": 0.0635, + "num_input_tokens_seen": 26357760, + "step": 20330 + }, + { + "epoch": 0.9935748662448391, + "grad_norm": 0.0012316630454733968, + "learning_rate": 0.1460656064689344, + "loss": 0.079, + "num_input_tokens_seen": 26364160, + "step": 20335 + }, + { + "epoch": 0.9938191679085334, + "grad_norm": 0.0007846902590245008, + "learning_rate": 0.14600672217760163, + "loss": 0.0907, + "num_input_tokens_seen": 26370432, + "step": 20340 + }, + { + "epoch": 0.9940634695722278, + "grad_norm": 0.003073090221732855, + "learning_rate": 0.14594783850208248, + "loss": 0.1239, + "num_input_tokens_seen": 26378272, + "step": 20345 + }, + { + "epoch": 0.9943077712359221, + "grad_norm": 0.001501662889495492, + "learning_rate": 0.14588895545145758, + "loss": 0.0984, + "num_input_tokens_seen": 26384800, + "step": 20350 + }, + { + "epoch": 0.9945520728996164, + "grad_norm": 0.0015304130502045155, + "learning_rate": 0.14583007303480738, + "loss": 0.1082, + "num_input_tokens_seen": 26391296, + "step": 20355 + }, + { + "epoch": 0.9947963745633108, + "grad_norm": 0.0009366472368128598, + "learning_rate": 0.14577119126121235, + "loss": 0.0958, + "num_input_tokens_seen": 26397824, + "step": 20360 + }, + { + "epoch": 0.9950406762270051, + "grad_norm": 0.0023777142632752657, + "learning_rate": 0.14571231013975272, + "loss": 0.1374, + "num_input_tokens_seen": 26404832, + "step": 20365 + }, + { + "epoch": 0.9952849778906995, + "grad_norm": 0.0007186130387708545, + "learning_rate": 0.1456534296795088, + "loss": 0.1023, + "num_input_tokens_seen": 26411712, + "step": 20370 + }, + { + "epoch": 0.9955292795543937, + "grad_norm": 0.0008455915376543999, + "learning_rate": 0.14559454988956066, + "loss": 0.0956, + "num_input_tokens_seen": 26417600, + "step": 20375 + }, + { + "epoch": 0.9957735812180881, + "grad_norm": 0.0007379049202427268, + "learning_rate": 0.1455356707789882, + "loss": 0.1043, + "num_input_tokens_seen": 26423648, + "step": 20380 + }, + { + "epoch": 0.9960178828817824, + "grad_norm": 0.001375784631818533, + "learning_rate": 0.14547679235687147, + "loss": 0.1051, + "num_input_tokens_seen": 26429632, + "step": 20385 + }, + { + "epoch": 0.9962621845454768, + "grad_norm": 0.0011764400405809283, + "learning_rate": 0.14541791463229023, + "loss": 0.1029, + "num_input_tokens_seen": 26436064, + "step": 20390 + }, + { + "epoch": 0.996506486209171, + "grad_norm": 0.0010301161091774702, + "learning_rate": 0.14535903761432406, + "loss": 0.0889, + "num_input_tokens_seen": 26443008, + "step": 20395 + }, + { + "epoch": 0.9967507878728654, + "grad_norm": 0.000988221843726933, + "learning_rate": 0.1453001613120527, + "loss": 0.0992, + "num_input_tokens_seen": 26449184, + "step": 20400 + }, + { + "epoch": 0.9967507878728654, + "eval_loss": 0.09594520181417465, + "eval_runtime": 401.9779, + "eval_samples_per_second": 90.515, + "eval_steps_per_second": 22.631, + "num_input_tokens_seen": 26449184, + "step": 20400 + }, + { + "epoch": 0.9969950895365598, + "grad_norm": 0.000730140833184123, + "learning_rate": 0.14524128573455547, + "loss": 0.0828, + "num_input_tokens_seen": 26455616, + "step": 20405 + }, + { + "epoch": 0.9972393912002541, + "grad_norm": 0.0007171963225118816, + "learning_rate": 0.14518241089091177, + "loss": 0.0754, + "num_input_tokens_seen": 26461920, + "step": 20410 + }, + { + "epoch": 0.9974836928639484, + "grad_norm": 0.00077079824404791, + "learning_rate": 0.1451235367902009, + "loss": 0.1071, + "num_input_tokens_seen": 26468192, + "step": 20415 + }, + { + "epoch": 0.9977279945276427, + "grad_norm": 0.0007430916302837431, + "learning_rate": 0.1450646634415019, + "loss": 0.0899, + "num_input_tokens_seen": 26474848, + "step": 20420 + }, + { + "epoch": 0.9979722961913371, + "grad_norm": 0.0015621059574186802, + "learning_rate": 0.1450057908538938, + "loss": 0.1081, + "num_input_tokens_seen": 26480960, + "step": 20425 + }, + { + "epoch": 0.9982165978550314, + "grad_norm": 0.0014769547851756215, + "learning_rate": 0.14494691903645557, + "loss": 0.1314, + "num_input_tokens_seen": 26486944, + "step": 20430 + }, + { + "epoch": 0.9984608995187257, + "grad_norm": 0.0016121063381433487, + "learning_rate": 0.14488804799826588, + "loss": 0.0833, + "num_input_tokens_seen": 26493728, + "step": 20435 + }, + { + "epoch": 0.99870520118242, + "grad_norm": 0.0007652232307009399, + "learning_rate": 0.14482917774840348, + "loss": 0.0755, + "num_input_tokens_seen": 26500640, + "step": 20440 + }, + { + "epoch": 0.9989495028461144, + "grad_norm": 0.0006797517999075353, + "learning_rate": 0.14477030829594684, + "loss": 0.0761, + "num_input_tokens_seen": 26507008, + "step": 20445 + }, + { + "epoch": 0.9991938045098087, + "grad_norm": 0.0006418574484996498, + "learning_rate": 0.14471143964997432, + "loss": 0.0802, + "num_input_tokens_seen": 26513312, + "step": 20450 + }, + { + "epoch": 0.999438106173503, + "grad_norm": 0.0013355695409700274, + "learning_rate": 0.14465257181956434, + "loss": 0.0602, + "num_input_tokens_seen": 26519776, + "step": 20455 + }, + { + "epoch": 0.9996824078371974, + "grad_norm": 0.0015310862800106406, + "learning_rate": 0.1445937048137949, + "loss": 0.1084, + "num_input_tokens_seen": 26525888, + "step": 20460 + }, + { + "epoch": 0.9999267095008917, + "grad_norm": 0.0005927713355049491, + "learning_rate": 0.14453483864174416, + "loss": 0.0656, + "num_input_tokens_seen": 26532384, + "step": 20465 + }, + { + "epoch": 1.0001954413309555, + "grad_norm": 0.002186753787100315, + "learning_rate": 0.14447597331249, + "loss": 0.0985, + "num_input_tokens_seen": 26538976, + "step": 20470 + }, + { + "epoch": 1.0004397429946499, + "grad_norm": 0.0009480072185397148, + "learning_rate": 0.1444171088351102, + "loss": 0.0786, + "num_input_tokens_seen": 26545568, + "step": 20475 + }, + { + "epoch": 1.000684044658344, + "grad_norm": 0.0010610298486426473, + "learning_rate": 0.14435824521868235, + "loss": 0.0796, + "num_input_tokens_seen": 26552448, + "step": 20480 + }, + { + "epoch": 1.0009283463220384, + "grad_norm": 0.0017356903990730643, + "learning_rate": 0.14429938247228397, + "loss": 0.0982, + "num_input_tokens_seen": 26558464, + "step": 20485 + }, + { + "epoch": 1.0011726479857328, + "grad_norm": 0.0005703876377083361, + "learning_rate": 0.14424052060499243, + "loss": 0.0811, + "num_input_tokens_seen": 26566528, + "step": 20490 + }, + { + "epoch": 1.0014169496494272, + "grad_norm": 0.0013428179081529379, + "learning_rate": 0.14418165962588506, + "loss": 0.1182, + "num_input_tokens_seen": 26572960, + "step": 20495 + }, + { + "epoch": 1.0016612513131213, + "grad_norm": 0.0011472615879029036, + "learning_rate": 0.1441227995440388, + "loss": 0.0896, + "num_input_tokens_seen": 26579296, + "step": 20500 + }, + { + "epoch": 1.0019055529768157, + "grad_norm": 0.001139275380410254, + "learning_rate": 0.14406394036853082, + "loss": 0.1073, + "num_input_tokens_seen": 26585664, + "step": 20505 + }, + { + "epoch": 1.00214985464051, + "grad_norm": 0.0013787519419565797, + "learning_rate": 0.14400508210843774, + "loss": 0.0901, + "num_input_tokens_seen": 26592800, + "step": 20510 + }, + { + "epoch": 1.0023941563042045, + "grad_norm": 0.0020843648817390203, + "learning_rate": 0.1439462247728364, + "loss": 0.1046, + "num_input_tokens_seen": 26599456, + "step": 20515 + }, + { + "epoch": 1.0026384579678989, + "grad_norm": 0.0011628777720034122, + "learning_rate": 0.14388736837080326, + "loss": 0.0767, + "num_input_tokens_seen": 26605760, + "step": 20520 + }, + { + "epoch": 1.002882759631593, + "grad_norm": 0.0013338197022676468, + "learning_rate": 0.14382851291141469, + "loss": 0.1186, + "num_input_tokens_seen": 26612544, + "step": 20525 + }, + { + "epoch": 1.0031270612952874, + "grad_norm": 0.000634790922049433, + "learning_rate": 0.14376965840374697, + "loss": 0.0909, + "num_input_tokens_seen": 26619488, + "step": 20530 + }, + { + "epoch": 1.0033713629589818, + "grad_norm": 0.0012371388729661703, + "learning_rate": 0.14371080485687632, + "loss": 0.089, + "num_input_tokens_seen": 26626144, + "step": 20535 + }, + { + "epoch": 1.0036156646226762, + "grad_norm": 0.0006532712723128498, + "learning_rate": 0.1436519522798785, + "loss": 0.0934, + "num_input_tokens_seen": 26632384, + "step": 20540 + }, + { + "epoch": 1.0038599662863703, + "grad_norm": 0.0007534011965617537, + "learning_rate": 0.14359310068182948, + "loss": 0.0782, + "num_input_tokens_seen": 26639168, + "step": 20545 + }, + { + "epoch": 1.0041042679500647, + "grad_norm": 0.0011416601482778788, + "learning_rate": 0.14353425007180484, + "loss": 0.0952, + "num_input_tokens_seen": 26646112, + "step": 20550 + }, + { + "epoch": 1.004348569613759, + "grad_norm": 0.0009490707889199257, + "learning_rate": 0.14347540045888005, + "loss": 0.0918, + "num_input_tokens_seen": 26652384, + "step": 20555 + }, + { + "epoch": 1.0045928712774534, + "grad_norm": 0.0012537569273263216, + "learning_rate": 0.14341655185213056, + "loss": 0.0853, + "num_input_tokens_seen": 26658912, + "step": 20560 + }, + { + "epoch": 1.0048371729411478, + "grad_norm": 0.0011037116637453437, + "learning_rate": 0.14335770426063144, + "loss": 0.0741, + "num_input_tokens_seen": 26665024, + "step": 20565 + }, + { + "epoch": 1.005081474604842, + "grad_norm": 0.00034392887027934194, + "learning_rate": 0.1432988576934578, + "loss": 0.0747, + "num_input_tokens_seen": 26671200, + "step": 20570 + }, + { + "epoch": 1.0053257762685364, + "grad_norm": 0.001014203648082912, + "learning_rate": 0.14324001215968457, + "loss": 0.0821, + "num_input_tokens_seen": 26677792, + "step": 20575 + }, + { + "epoch": 1.0055700779322307, + "grad_norm": 0.0015979910967871547, + "learning_rate": 0.14318116766838637, + "loss": 0.1186, + "num_input_tokens_seen": 26684512, + "step": 20580 + }, + { + "epoch": 1.0058143795959251, + "grad_norm": 0.0006908914656378329, + "learning_rate": 0.14312232422863788, + "loss": 0.0853, + "num_input_tokens_seen": 26690880, + "step": 20585 + }, + { + "epoch": 1.0060586812596193, + "grad_norm": 0.0012840059353038669, + "learning_rate": 0.14306348184951334, + "loss": 0.0765, + "num_input_tokens_seen": 26697504, + "step": 20590 + }, + { + "epoch": 1.0063029829233137, + "grad_norm": 0.0016314949607476592, + "learning_rate": 0.1430046405400871, + "loss": 0.0863, + "num_input_tokens_seen": 26703808, + "step": 20595 + }, + { + "epoch": 1.006547284587008, + "grad_norm": 0.0014612477971240878, + "learning_rate": 0.14294580030943324, + "loss": 0.1161, + "num_input_tokens_seen": 26710048, + "step": 20600 + }, + { + "epoch": 1.006547284587008, + "eval_loss": 0.09938289225101471, + "eval_runtime": 402.3605, + "eval_samples_per_second": 90.429, + "eval_steps_per_second": 22.609, + "num_input_tokens_seen": 26710048, + "step": 20600 + }, + { + "epoch": 1.0067915862507024, + "grad_norm": 0.0013994949404150248, + "learning_rate": 0.14288696116662553, + "loss": 0.0987, + "num_input_tokens_seen": 26716544, + "step": 20605 + }, + { + "epoch": 1.0070358879143968, + "grad_norm": 0.0010188705055043101, + "learning_rate": 0.1428281231207378, + "loss": 0.0939, + "num_input_tokens_seen": 26722880, + "step": 20610 + }, + { + "epoch": 1.007280189578091, + "grad_norm": 0.0008885830175131559, + "learning_rate": 0.1427692861808437, + "loss": 0.0905, + "num_input_tokens_seen": 26729600, + "step": 20615 + }, + { + "epoch": 1.0075244912417853, + "grad_norm": 0.0009953760309144855, + "learning_rate": 0.1427104503560165, + "loss": 0.1209, + "num_input_tokens_seen": 26735968, + "step": 20620 + }, + { + "epoch": 1.0077687929054797, + "grad_norm": 0.0007720266003161669, + "learning_rate": 0.14265161565532947, + "loss": 0.0852, + "num_input_tokens_seen": 26742304, + "step": 20625 + }, + { + "epoch": 1.008013094569174, + "grad_norm": 0.0008837715722620487, + "learning_rate": 0.14259278208785564, + "loss": 0.0575, + "num_input_tokens_seen": 26749024, + "step": 20630 + }, + { + "epoch": 1.0082573962328683, + "grad_norm": 0.0012705313274636865, + "learning_rate": 0.14253394966266789, + "loss": 0.0911, + "num_input_tokens_seen": 26755328, + "step": 20635 + }, + { + "epoch": 1.0085016978965626, + "grad_norm": 0.0011299996403977275, + "learning_rate": 0.14247511838883894, + "loss": 0.0793, + "num_input_tokens_seen": 26761760, + "step": 20640 + }, + { + "epoch": 1.008745999560257, + "grad_norm": 0.0009928969666361809, + "learning_rate": 0.14241628827544126, + "loss": 0.1077, + "num_input_tokens_seen": 26767872, + "step": 20645 + }, + { + "epoch": 1.0089903012239514, + "grad_norm": 0.0005016888026148081, + "learning_rate": 0.14235745933154723, + "loss": 0.0923, + "num_input_tokens_seen": 26774112, + "step": 20650 + }, + { + "epoch": 1.0092346028876458, + "grad_norm": 0.001409450895152986, + "learning_rate": 0.14229863156622907, + "loss": 0.0882, + "num_input_tokens_seen": 26780832, + "step": 20655 + }, + { + "epoch": 1.00947890455134, + "grad_norm": 0.0011605179170146585, + "learning_rate": 0.14223980498855868, + "loss": 0.0864, + "num_input_tokens_seen": 26787264, + "step": 20660 + }, + { + "epoch": 1.0097232062150343, + "grad_norm": 0.0007030564011074603, + "learning_rate": 0.14218097960760792, + "loss": 0.0846, + "num_input_tokens_seen": 26793312, + "step": 20665 + }, + { + "epoch": 1.0099675078787287, + "grad_norm": 0.000910881906747818, + "learning_rate": 0.1421221554324483, + "loss": 0.0919, + "num_input_tokens_seen": 26799840, + "step": 20670 + }, + { + "epoch": 1.010211809542423, + "grad_norm": 0.001189140835776925, + "learning_rate": 0.1420633324721513, + "loss": 0.0877, + "num_input_tokens_seen": 26806816, + "step": 20675 + }, + { + "epoch": 1.0104561112061172, + "grad_norm": 0.0012368238531053066, + "learning_rate": 0.14200451073578824, + "loss": 0.1142, + "num_input_tokens_seen": 26813344, + "step": 20680 + }, + { + "epoch": 1.0107004128698116, + "grad_norm": 0.0008418634533882141, + "learning_rate": 0.14194569023243003, + "loss": 0.0947, + "num_input_tokens_seen": 26819616, + "step": 20685 + }, + { + "epoch": 1.010944714533506, + "grad_norm": 0.001124703325331211, + "learning_rate": 0.14188687097114766, + "loss": 0.1156, + "num_input_tokens_seen": 26826272, + "step": 20690 + }, + { + "epoch": 1.0111890161972004, + "grad_norm": 0.0020638855639845133, + "learning_rate": 0.14182805296101172, + "loss": 0.103, + "num_input_tokens_seen": 26832704, + "step": 20695 + }, + { + "epoch": 1.0114333178608947, + "grad_norm": 0.0009343070560134947, + "learning_rate": 0.14176923621109272, + "loss": 0.0766, + "num_input_tokens_seen": 26839424, + "step": 20700 + }, + { + "epoch": 1.011677619524589, + "grad_norm": 0.0015224473318085074, + "learning_rate": 0.14171042073046097, + "loss": 0.112, + "num_input_tokens_seen": 26845568, + "step": 20705 + }, + { + "epoch": 1.0119219211882833, + "grad_norm": 0.0011195959523320198, + "learning_rate": 0.14165160652818642, + "loss": 0.0886, + "num_input_tokens_seen": 26852032, + "step": 20710 + }, + { + "epoch": 1.0121662228519777, + "grad_norm": 0.0010892233112826943, + "learning_rate": 0.14159279361333907, + "loss": 0.0811, + "num_input_tokens_seen": 26858304, + "step": 20715 + }, + { + "epoch": 1.012410524515672, + "grad_norm": 0.0014277203008532524, + "learning_rate": 0.14153398199498868, + "loss": 0.1009, + "num_input_tokens_seen": 26865504, + "step": 20720 + }, + { + "epoch": 1.0126548261793662, + "grad_norm": 0.0010732711525633931, + "learning_rate": 0.14147517168220458, + "loss": 0.0821, + "num_input_tokens_seen": 26872160, + "step": 20725 + }, + { + "epoch": 1.0128991278430606, + "grad_norm": 0.001859849551692605, + "learning_rate": 0.14141636268405616, + "loss": 0.1247, + "num_input_tokens_seen": 26878368, + "step": 20730 + }, + { + "epoch": 1.013143429506755, + "grad_norm": 0.0014696632279083133, + "learning_rate": 0.14135755500961253, + "loss": 0.0949, + "num_input_tokens_seen": 26885088, + "step": 20735 + }, + { + "epoch": 1.0133877311704493, + "grad_norm": 0.0012999975588172674, + "learning_rate": 0.14129874866794245, + "loss": 0.103, + "num_input_tokens_seen": 26891584, + "step": 20740 + }, + { + "epoch": 1.0136320328341435, + "grad_norm": 0.0011912272311747074, + "learning_rate": 0.14123994366811476, + "loss": 0.097, + "num_input_tokens_seen": 26898112, + "step": 20745 + }, + { + "epoch": 1.0138763344978379, + "grad_norm": 0.0013300852151587605, + "learning_rate": 0.14118114001919774, + "loss": 0.077, + "num_input_tokens_seen": 26905088, + "step": 20750 + }, + { + "epoch": 1.0141206361615323, + "grad_norm": 0.0012279768707230687, + "learning_rate": 0.14112233773025978, + "loss": 0.0679, + "num_input_tokens_seen": 26911968, + "step": 20755 + }, + { + "epoch": 1.0143649378252266, + "grad_norm": 0.0011865205597132444, + "learning_rate": 0.14106353681036896, + "loss": 0.0995, + "num_input_tokens_seen": 26918272, + "step": 20760 + }, + { + "epoch": 1.014609239488921, + "grad_norm": 0.0006208817940205336, + "learning_rate": 0.14100473726859303, + "loss": 0.1266, + "num_input_tokens_seen": 26924672, + "step": 20765 + }, + { + "epoch": 1.0148535411526152, + "grad_norm": 0.000884358596522361, + "learning_rate": 0.14094593911399964, + "loss": 0.1057, + "num_input_tokens_seen": 26930688, + "step": 20770 + }, + { + "epoch": 1.0150978428163095, + "grad_norm": 0.0006318718078546226, + "learning_rate": 0.14088714235565625, + "loss": 0.0847, + "num_input_tokens_seen": 26937024, + "step": 20775 + }, + { + "epoch": 1.015342144480004, + "grad_norm": 0.000848635972943157, + "learning_rate": 0.14082834700263, + "loss": 0.0738, + "num_input_tokens_seen": 26943616, + "step": 20780 + }, + { + "epoch": 1.0155864461436983, + "grad_norm": 0.0008753227884881198, + "learning_rate": 0.14076955306398795, + "loss": 0.0954, + "num_input_tokens_seen": 26949952, + "step": 20785 + }, + { + "epoch": 1.0158307478073925, + "grad_norm": 0.0014150083297863603, + "learning_rate": 0.14071076054879675, + "loss": 0.1189, + "num_input_tokens_seen": 26956192, + "step": 20790 + }, + { + "epoch": 1.0160750494710868, + "grad_norm": 0.001134058111347258, + "learning_rate": 0.14065196946612302, + "loss": 0.0887, + "num_input_tokens_seen": 26962464, + "step": 20795 + }, + { + "epoch": 1.0163193511347812, + "grad_norm": 0.0015585428336635232, + "learning_rate": 0.1405931798250331, + "loss": 0.1113, + "num_input_tokens_seen": 26968800, + "step": 20800 + }, + { + "epoch": 1.0163193511347812, + "eval_loss": 0.09433497488498688, + "eval_runtime": 402.7263, + "eval_samples_per_second": 90.347, + "eval_steps_per_second": 22.589, + "num_input_tokens_seen": 26968800, + "step": 20800 + }, + { + "epoch": 1.0165636527984756, + "grad_norm": 0.0013650258770212531, + "learning_rate": 0.14053439163459308, + "loss": 0.0954, + "num_input_tokens_seen": 26975936, + "step": 20805 + }, + { + "epoch": 1.01680795446217, + "grad_norm": 0.001086244941689074, + "learning_rate": 0.14047560490386876, + "loss": 0.0902, + "num_input_tokens_seen": 26982656, + "step": 20810 + }, + { + "epoch": 1.0170522561258641, + "grad_norm": 0.0012559453025460243, + "learning_rate": 0.14041681964192593, + "loss": 0.0846, + "num_input_tokens_seen": 26989760, + "step": 20815 + }, + { + "epoch": 1.0172965577895585, + "grad_norm": 0.0007597898365929723, + "learning_rate": 0.14035803585782988, + "loss": 0.08, + "num_input_tokens_seen": 26996576, + "step": 20820 + }, + { + "epoch": 1.017540859453253, + "grad_norm": 0.000990328611806035, + "learning_rate": 0.14029925356064593, + "loss": 0.0691, + "num_input_tokens_seen": 27003232, + "step": 20825 + }, + { + "epoch": 1.0177851611169473, + "grad_norm": 0.0011675377609208226, + "learning_rate": 0.1402404727594389, + "loss": 0.0775, + "num_input_tokens_seen": 27009312, + "step": 20830 + }, + { + "epoch": 1.0180294627806414, + "grad_norm": 0.0012729184236377478, + "learning_rate": 0.1401816934632737, + "loss": 0.0753, + "num_input_tokens_seen": 27015552, + "step": 20835 + }, + { + "epoch": 1.0182737644443358, + "grad_norm": 0.0009322831756435335, + "learning_rate": 0.1401229156812147, + "loss": 0.0772, + "num_input_tokens_seen": 27022336, + "step": 20840 + }, + { + "epoch": 1.0185180661080302, + "grad_norm": 0.0009345178259536624, + "learning_rate": 0.14006413942232626, + "loss": 0.0709, + "num_input_tokens_seen": 27028800, + "step": 20845 + }, + { + "epoch": 1.0187623677717246, + "grad_norm": 0.0007351326639764011, + "learning_rate": 0.14000536469567235, + "loss": 0.0716, + "num_input_tokens_seen": 27035584, + "step": 20850 + }, + { + "epoch": 1.019006669435419, + "grad_norm": 0.0016152994940057397, + "learning_rate": 0.13994659151031685, + "loss": 0.1009, + "num_input_tokens_seen": 27042400, + "step": 20855 + }, + { + "epoch": 1.0192509710991131, + "grad_norm": 0.001337930210866034, + "learning_rate": 0.13988781987532323, + "loss": 0.1253, + "num_input_tokens_seen": 27048704, + "step": 20860 + }, + { + "epoch": 1.0194952727628075, + "grad_norm": 0.0008672375697642565, + "learning_rate": 0.1398290497997549, + "loss": 0.0769, + "num_input_tokens_seen": 27055296, + "step": 20865 + }, + { + "epoch": 1.0197395744265019, + "grad_norm": 0.0012540515745058656, + "learning_rate": 0.13977028129267488, + "loss": 0.1066, + "num_input_tokens_seen": 27061824, + "step": 20870 + }, + { + "epoch": 1.0199838760901963, + "grad_norm": 0.0012696873163804412, + "learning_rate": 0.13971151436314605, + "loss": 0.0887, + "num_input_tokens_seen": 27068352, + "step": 20875 + }, + { + "epoch": 1.0202281777538904, + "grad_norm": 0.0007751485682092607, + "learning_rate": 0.13965274902023103, + "loss": 0.0868, + "num_input_tokens_seen": 27074816, + "step": 20880 + }, + { + "epoch": 1.0204724794175848, + "grad_norm": 0.0010521267540752888, + "learning_rate": 0.13959398527299208, + "loss": 0.1038, + "num_input_tokens_seen": 27081504, + "step": 20885 + }, + { + "epoch": 1.0207167810812792, + "grad_norm": 0.0004944460233673453, + "learning_rate": 0.13953522313049138, + "loss": 0.0683, + "num_input_tokens_seen": 27087744, + "step": 20890 + }, + { + "epoch": 1.0209610827449735, + "grad_norm": 0.0010819894960150123, + "learning_rate": 0.13947646260179083, + "loss": 0.0967, + "num_input_tokens_seen": 27094464, + "step": 20895 + }, + { + "epoch": 1.021205384408668, + "grad_norm": 0.009754388593137264, + "learning_rate": 0.13941770369595194, + "loss": 0.111, + "num_input_tokens_seen": 27100512, + "step": 20900 + }, + { + "epoch": 1.021449686072362, + "grad_norm": 0.001250391243956983, + "learning_rate": 0.1393589464220362, + "loss": 0.1158, + "num_input_tokens_seen": 27107008, + "step": 20905 + }, + { + "epoch": 1.0216939877360565, + "grad_norm": 0.0012137795565649867, + "learning_rate": 0.13930019078910455, + "loss": 0.1022, + "num_input_tokens_seen": 27113472, + "step": 20910 + }, + { + "epoch": 1.0219382893997508, + "grad_norm": 0.0011166553013026714, + "learning_rate": 0.139241436806218, + "loss": 0.1077, + "num_input_tokens_seen": 27119808, + "step": 20915 + }, + { + "epoch": 1.0221825910634452, + "grad_norm": 0.00216299993917346, + "learning_rate": 0.13918268448243712, + "loss": 0.0844, + "num_input_tokens_seen": 27125920, + "step": 20920 + }, + { + "epoch": 1.0224268927271394, + "grad_norm": 0.0008488960447721183, + "learning_rate": 0.13912393382682217, + "loss": 0.0851, + "num_input_tokens_seen": 27132960, + "step": 20925 + }, + { + "epoch": 1.0226711943908338, + "grad_norm": 0.0014313937863335013, + "learning_rate": 0.1390651848484333, + "loss": 0.0937, + "num_input_tokens_seen": 27140000, + "step": 20930 + }, + { + "epoch": 1.0229154960545281, + "grad_norm": 0.0011079367250204086, + "learning_rate": 0.1390064375563304, + "loss": 0.0805, + "num_input_tokens_seen": 27146080, + "step": 20935 + }, + { + "epoch": 1.0231597977182225, + "grad_norm": 0.001337229274213314, + "learning_rate": 0.13894769195957293, + "loss": 0.095, + "num_input_tokens_seen": 27152416, + "step": 20940 + }, + { + "epoch": 1.023404099381917, + "grad_norm": 0.0010114923352375627, + "learning_rate": 0.13888894806722032, + "loss": 0.0886, + "num_input_tokens_seen": 27159200, + "step": 20945 + }, + { + "epoch": 1.023648401045611, + "grad_norm": 0.0009811188792809844, + "learning_rate": 0.1388302058883315, + "loss": 0.089, + "num_input_tokens_seen": 27165472, + "step": 20950 + }, + { + "epoch": 1.0238927027093054, + "grad_norm": 0.000931074027903378, + "learning_rate": 0.13877146543196528, + "loss": 0.0964, + "num_input_tokens_seen": 27172064, + "step": 20955 + }, + { + "epoch": 1.0241370043729998, + "grad_norm": 0.0008057938539423048, + "learning_rate": 0.13871272670718027, + "loss": 0.1027, + "num_input_tokens_seen": 27178432, + "step": 20960 + }, + { + "epoch": 1.0243813060366942, + "grad_norm": 0.00116734707262367, + "learning_rate": 0.13865398972303455, + "loss": 0.088, + "num_input_tokens_seen": 27184992, + "step": 20965 + }, + { + "epoch": 1.0246256077003884, + "grad_norm": 0.0012190502602607012, + "learning_rate": 0.13859525448858623, + "loss": 0.0795, + "num_input_tokens_seen": 27191712, + "step": 20970 + }, + { + "epoch": 1.0248699093640827, + "grad_norm": 0.0010124478721991181, + "learning_rate": 0.13853652101289304, + "loss": 0.081, + "num_input_tokens_seen": 27198048, + "step": 20975 + }, + { + "epoch": 1.0251142110277771, + "grad_norm": 0.0013032964197918773, + "learning_rate": 0.13847778930501234, + "loss": 0.0841, + "num_input_tokens_seen": 27204384, + "step": 20980 + }, + { + "epoch": 1.0253585126914715, + "grad_norm": 0.0009032879606820643, + "learning_rate": 0.1384190593740013, + "loss": 0.1087, + "num_input_tokens_seen": 27210592, + "step": 20985 + }, + { + "epoch": 1.0256028143551656, + "grad_norm": 0.0015317421639338136, + "learning_rate": 0.13836033122891686, + "loss": 0.0963, + "num_input_tokens_seen": 27216960, + "step": 20990 + }, + { + "epoch": 1.02584711601886, + "grad_norm": 0.001095977844670415, + "learning_rate": 0.1383016048788156, + "loss": 0.1265, + "num_input_tokens_seen": 27223552, + "step": 20995 + }, + { + "epoch": 1.0260914176825544, + "grad_norm": 0.0012844977900385857, + "learning_rate": 0.13824288033275392, + "loss": 0.0776, + "num_input_tokens_seen": 27230240, + "step": 21000 + }, + { + "epoch": 1.0260914176825544, + "eval_loss": 0.1055796816945076, + "eval_runtime": 402.1425, + "eval_samples_per_second": 90.478, + "eval_steps_per_second": 22.621, + "num_input_tokens_seen": 27230240, + "step": 21000 + }, + { + "epoch": 1.0263357193462488, + "grad_norm": 0.0008730414556339383, + "learning_rate": 0.1381841575997878, + "loss": 0.0884, + "num_input_tokens_seen": 27236704, + "step": 21005 + }, + { + "epoch": 1.0265800210099432, + "grad_norm": 0.0009742409456521273, + "learning_rate": 0.13812543668897306, + "loss": 0.0951, + "num_input_tokens_seen": 27242688, + "step": 21010 + }, + { + "epoch": 1.0268243226736373, + "grad_norm": 0.002033266704529524, + "learning_rate": 0.13806671760936526, + "loss": 0.1033, + "num_input_tokens_seen": 27249408, + "step": 21015 + }, + { + "epoch": 1.0270686243373317, + "grad_norm": 0.0008926842710934579, + "learning_rate": 0.13800800037001956, + "loss": 0.0812, + "num_input_tokens_seen": 27255872, + "step": 21020 + }, + { + "epoch": 1.027312926001026, + "grad_norm": 0.001288104336708784, + "learning_rate": 0.13794928497999087, + "loss": 0.0623, + "num_input_tokens_seen": 27262240, + "step": 21025 + }, + { + "epoch": 1.0275572276647205, + "grad_norm": 0.0011489654425531626, + "learning_rate": 0.1378905714483339, + "loss": 0.1038, + "num_input_tokens_seen": 27268384, + "step": 21030 + }, + { + "epoch": 1.0278015293284146, + "grad_norm": 0.0012247991980984807, + "learning_rate": 0.13783185978410295, + "loss": 0.0667, + "num_input_tokens_seen": 27274912, + "step": 21035 + }, + { + "epoch": 1.028045830992109, + "grad_norm": 0.00159212586004287, + "learning_rate": 0.13777314999635218, + "loss": 0.1005, + "num_input_tokens_seen": 27281280, + "step": 21040 + }, + { + "epoch": 1.0282901326558034, + "grad_norm": 0.0014268578961491585, + "learning_rate": 0.1377144420941353, + "loss": 0.0913, + "num_input_tokens_seen": 27287520, + "step": 21045 + }, + { + "epoch": 1.0285344343194978, + "grad_norm": 0.0015036859549582005, + "learning_rate": 0.13765573608650586, + "loss": 0.1319, + "num_input_tokens_seen": 27294464, + "step": 21050 + }, + { + "epoch": 1.0287787359831921, + "grad_norm": 0.0013794702244922519, + "learning_rate": 0.13759703198251702, + "loss": 0.0954, + "num_input_tokens_seen": 27301024, + "step": 21055 + }, + { + "epoch": 1.0290230376468863, + "grad_norm": 0.0014114645309746265, + "learning_rate": 0.13753832979122174, + "loss": 0.1206, + "num_input_tokens_seen": 27307808, + "step": 21060 + }, + { + "epoch": 1.0292673393105807, + "grad_norm": 0.001893583103083074, + "learning_rate": 0.13747962952167264, + "loss": 0.0983, + "num_input_tokens_seen": 27314272, + "step": 21065 + }, + { + "epoch": 1.029511640974275, + "grad_norm": 0.0010860050097107887, + "learning_rate": 0.13742093118292192, + "loss": 0.1005, + "num_input_tokens_seen": 27320992, + "step": 21070 + }, + { + "epoch": 1.0297559426379694, + "grad_norm": 0.0009358865208923817, + "learning_rate": 0.13736223478402174, + "loss": 0.097, + "num_input_tokens_seen": 27327488, + "step": 21075 + }, + { + "epoch": 1.0300002443016636, + "grad_norm": 0.0006481375312432647, + "learning_rate": 0.1373035403340238, + "loss": 0.0846, + "num_input_tokens_seen": 27333728, + "step": 21080 + }, + { + "epoch": 1.030244545965358, + "grad_norm": 0.0012126066721975803, + "learning_rate": 0.13724484784197943, + "loss": 0.0817, + "num_input_tokens_seen": 27340192, + "step": 21085 + }, + { + "epoch": 1.0304888476290524, + "grad_norm": 0.0008379507926292717, + "learning_rate": 0.13718615731693987, + "loss": 0.1208, + "num_input_tokens_seen": 27346528, + "step": 21090 + }, + { + "epoch": 1.0307331492927467, + "grad_norm": 0.0010841719340533018, + "learning_rate": 0.13712746876795587, + "loss": 0.1079, + "num_input_tokens_seen": 27352832, + "step": 21095 + }, + { + "epoch": 1.030977450956441, + "grad_norm": 0.0008829680155031383, + "learning_rate": 0.13706878220407792, + "loss": 0.0934, + "num_input_tokens_seen": 27359328, + "step": 21100 + }, + { + "epoch": 1.0312217526201353, + "grad_norm": 0.0007203482091426849, + "learning_rate": 0.13701009763435631, + "loss": 0.079, + "num_input_tokens_seen": 27366080, + "step": 21105 + }, + { + "epoch": 1.0314660542838296, + "grad_norm": 0.0007952941232360899, + "learning_rate": 0.13695141506784084, + "loss": 0.0995, + "num_input_tokens_seen": 27372736, + "step": 21110 + }, + { + "epoch": 1.031710355947524, + "grad_norm": 0.0011358850169926882, + "learning_rate": 0.13689273451358114, + "loss": 0.1019, + "num_input_tokens_seen": 27379328, + "step": 21115 + }, + { + "epoch": 1.0319546576112184, + "grad_norm": 0.003053368767723441, + "learning_rate": 0.13683405598062653, + "loss": 0.1304, + "num_input_tokens_seen": 27385888, + "step": 21120 + }, + { + "epoch": 1.0321989592749126, + "grad_norm": 0.0014941069530323148, + "learning_rate": 0.1367753794780259, + "loss": 0.0916, + "num_input_tokens_seen": 27391904, + "step": 21125 + }, + { + "epoch": 1.032443260938607, + "grad_norm": 0.0027936934493482113, + "learning_rate": 0.13671670501482802, + "loss": 0.1055, + "num_input_tokens_seen": 27398176, + "step": 21130 + }, + { + "epoch": 1.0326875626023013, + "grad_norm": 0.0015457833651453257, + "learning_rate": 0.1366580326000811, + "loss": 0.0978, + "num_input_tokens_seen": 27404512, + "step": 21135 + }, + { + "epoch": 1.0329318642659957, + "grad_norm": 0.0013659369433298707, + "learning_rate": 0.1365993622428332, + "loss": 0.1073, + "num_input_tokens_seen": 27411136, + "step": 21140 + }, + { + "epoch": 1.03317616592969, + "grad_norm": 0.0007714607054367661, + "learning_rate": 0.13654069395213211, + "loss": 0.1066, + "num_input_tokens_seen": 27417344, + "step": 21145 + }, + { + "epoch": 1.0334204675933842, + "grad_norm": 0.0010756602277979255, + "learning_rate": 0.13648202773702509, + "loss": 0.0906, + "num_input_tokens_seen": 27424064, + "step": 21150 + }, + { + "epoch": 1.0336647692570786, + "grad_norm": 0.0007315311231650412, + "learning_rate": 0.13642336360655927, + "loss": 0.0921, + "num_input_tokens_seen": 27430304, + "step": 21155 + }, + { + "epoch": 1.033909070920773, + "grad_norm": 0.0011198463616892695, + "learning_rate": 0.13636470156978145, + "loss": 0.1386, + "num_input_tokens_seen": 27436736, + "step": 21160 + }, + { + "epoch": 1.0341533725844674, + "grad_norm": 0.0010656729573383927, + "learning_rate": 0.13630604163573798, + "loss": 0.1053, + "num_input_tokens_seen": 27443232, + "step": 21165 + }, + { + "epoch": 1.0343976742481615, + "grad_norm": 0.0009600571356713772, + "learning_rate": 0.13624738381347495, + "loss": 0.0714, + "num_input_tokens_seen": 27449600, + "step": 21170 + }, + { + "epoch": 1.034641975911856, + "grad_norm": 0.0008991071372292936, + "learning_rate": 0.1361887281120382, + "loss": 0.0896, + "num_input_tokens_seen": 27456032, + "step": 21175 + }, + { + "epoch": 1.0348862775755503, + "grad_norm": 0.0007890230044722557, + "learning_rate": 0.13613007454047307, + "loss": 0.0805, + "num_input_tokens_seen": 27462432, + "step": 21180 + }, + { + "epoch": 1.0351305792392447, + "grad_norm": 0.001027237274684012, + "learning_rate": 0.13607142310782486, + "loss": 0.1064, + "num_input_tokens_seen": 27469440, + "step": 21185 + }, + { + "epoch": 1.035374880902939, + "grad_norm": 0.0006541185430251062, + "learning_rate": 0.13601277382313814, + "loss": 0.0807, + "num_input_tokens_seen": 27476096, + "step": 21190 + }, + { + "epoch": 1.0356191825666332, + "grad_norm": 0.000801075017079711, + "learning_rate": 0.1359541266954575, + "loss": 0.0693, + "num_input_tokens_seen": 27482560, + "step": 21195 + }, + { + "epoch": 1.0358634842303276, + "grad_norm": 0.0008625138434581459, + "learning_rate": 0.13589548173382707, + "loss": 0.07, + "num_input_tokens_seen": 27489152, + "step": 21200 + }, + { + "epoch": 1.0358634842303276, + "eval_loss": 0.09864918887615204, + "eval_runtime": 402.9034, + "eval_samples_per_second": 90.307, + "eval_steps_per_second": 22.579, + "num_input_tokens_seen": 27489152, + "step": 21200 + }, + { + "epoch": 1.036107785894022, + "grad_norm": 0.0005832379683852196, + "learning_rate": 0.1358368389472906, + "loss": 0.0869, + "num_input_tokens_seen": 27495008, + "step": 21205 + }, + { + "epoch": 1.0363520875577164, + "grad_norm": 0.0011426935670897365, + "learning_rate": 0.13577819834489155, + "loss": 0.0881, + "num_input_tokens_seen": 27501568, + "step": 21210 + }, + { + "epoch": 1.0365963892214105, + "grad_norm": 0.0007401816546916962, + "learning_rate": 0.135719559935673, + "loss": 0.1063, + "num_input_tokens_seen": 27508032, + "step": 21215 + }, + { + "epoch": 1.0368406908851049, + "grad_norm": 0.0017990939086303115, + "learning_rate": 0.13566092372867775, + "loss": 0.071, + "num_input_tokens_seen": 27514656, + "step": 21220 + }, + { + "epoch": 1.0370849925487993, + "grad_norm": 0.0014070946490392089, + "learning_rate": 0.13560228973294833, + "loss": 0.1054, + "num_input_tokens_seen": 27520608, + "step": 21225 + }, + { + "epoch": 1.0373292942124936, + "grad_norm": 0.000940449652262032, + "learning_rate": 0.13554365795752668, + "loss": 0.0778, + "num_input_tokens_seen": 27527456, + "step": 21230 + }, + { + "epoch": 1.037573595876188, + "grad_norm": 0.0010203662095591426, + "learning_rate": 0.1354850284114547, + "loss": 0.0615, + "num_input_tokens_seen": 27533888, + "step": 21235 + }, + { + "epoch": 1.0378178975398822, + "grad_norm": 0.00059178052470088, + "learning_rate": 0.13542640110377374, + "loss": 0.0865, + "num_input_tokens_seen": 27540544, + "step": 21240 + }, + { + "epoch": 1.0380621992035766, + "grad_norm": 0.0009657465270720422, + "learning_rate": 0.13536777604352487, + "loss": 0.0891, + "num_input_tokens_seen": 27546816, + "step": 21245 + }, + { + "epoch": 1.038306500867271, + "grad_norm": 0.0015366475563496351, + "learning_rate": 0.13530915323974887, + "loss": 0.1245, + "num_input_tokens_seen": 27553408, + "step": 21250 + }, + { + "epoch": 1.0385508025309653, + "grad_norm": 0.001082811621017754, + "learning_rate": 0.13525053270148596, + "loss": 0.0767, + "num_input_tokens_seen": 27559808, + "step": 21255 + }, + { + "epoch": 1.0387951041946595, + "grad_norm": 0.0016546017723158002, + "learning_rate": 0.13519191443777628, + "loss": 0.0917, + "num_input_tokens_seen": 27566080, + "step": 21260 + }, + { + "epoch": 1.0390394058583539, + "grad_norm": 0.0006919643492437899, + "learning_rate": 0.13513329845765953, + "loss": 0.0751, + "num_input_tokens_seen": 27572672, + "step": 21265 + }, + { + "epoch": 1.0392837075220482, + "grad_norm": 0.0006633157026953995, + "learning_rate": 0.13507468477017495, + "loss": 0.0702, + "num_input_tokens_seen": 27579136, + "step": 21270 + }, + { + "epoch": 1.0395280091857426, + "grad_norm": 0.0009227843256667256, + "learning_rate": 0.13501607338436153, + "loss": 0.0908, + "num_input_tokens_seen": 27585632, + "step": 21275 + }, + { + "epoch": 1.039772310849437, + "grad_norm": 0.0012821357231587172, + "learning_rate": 0.13495746430925798, + "loss": 0.0936, + "num_input_tokens_seen": 27591904, + "step": 21280 + }, + { + "epoch": 1.0400166125131312, + "grad_norm": 0.0008243247284553945, + "learning_rate": 0.13489885755390238, + "loss": 0.0654, + "num_input_tokens_seen": 27598720, + "step": 21285 + }, + { + "epoch": 1.0402609141768255, + "grad_norm": 0.0012429701164364815, + "learning_rate": 0.13484025312733275, + "loss": 0.0873, + "num_input_tokens_seen": 27605216, + "step": 21290 + }, + { + "epoch": 1.04050521584052, + "grad_norm": 0.0008658512379042804, + "learning_rate": 0.13478165103858658, + "loss": 0.0604, + "num_input_tokens_seen": 27612192, + "step": 21295 + }, + { + "epoch": 1.0407495175042143, + "grad_norm": 0.0010695260716602206, + "learning_rate": 0.13472305129670106, + "loss": 0.0885, + "num_input_tokens_seen": 27618720, + "step": 21300 + }, + { + "epoch": 1.0409938191679085, + "grad_norm": 0.001006843289360404, + "learning_rate": 0.13466445391071305, + "loss": 0.0641, + "num_input_tokens_seen": 27624832, + "step": 21305 + }, + { + "epoch": 1.0412381208316028, + "grad_norm": 0.0011336390161886811, + "learning_rate": 0.13460585888965895, + "loss": 0.1073, + "num_input_tokens_seen": 27630880, + "step": 21310 + }, + { + "epoch": 1.0414824224952972, + "grad_norm": 0.0014411498559638858, + "learning_rate": 0.13454726624257482, + "loss": 0.0874, + "num_input_tokens_seen": 27637664, + "step": 21315 + }, + { + "epoch": 1.0417267241589916, + "grad_norm": 0.0018521333113312721, + "learning_rate": 0.1344886759784965, + "loss": 0.0779, + "num_input_tokens_seen": 27644224, + "step": 21320 + }, + { + "epoch": 1.0419710258226857, + "grad_norm": 0.0013456186279654503, + "learning_rate": 0.13443008810645923, + "loss": 0.1167, + "num_input_tokens_seen": 27650304, + "step": 21325 + }, + { + "epoch": 1.0422153274863801, + "grad_norm": 0.0010382996406406164, + "learning_rate": 0.13437150263549807, + "loss": 0.1117, + "num_input_tokens_seen": 27657024, + "step": 21330 + }, + { + "epoch": 1.0424596291500745, + "grad_norm": 0.0008089079055935144, + "learning_rate": 0.13431291957464755, + "loss": 0.0813, + "num_input_tokens_seen": 27663360, + "step": 21335 + }, + { + "epoch": 1.0427039308137689, + "grad_norm": 0.0011522357817739248, + "learning_rate": 0.13425433893294197, + "loss": 0.0877, + "num_input_tokens_seen": 27670304, + "step": 21340 + }, + { + "epoch": 1.0429482324774633, + "grad_norm": 0.0008403757237829268, + "learning_rate": 0.13419576071941525, + "loss": 0.0981, + "num_input_tokens_seen": 27676416, + "step": 21345 + }, + { + "epoch": 1.0431925341411574, + "grad_norm": 0.0007292656810022891, + "learning_rate": 0.1341371849431008, + "loss": 0.0835, + "num_input_tokens_seen": 27683008, + "step": 21350 + }, + { + "epoch": 1.0434368358048518, + "grad_norm": 0.0007884882506914437, + "learning_rate": 0.13407861161303178, + "loss": 0.0826, + "num_input_tokens_seen": 27689536, + "step": 21355 + }, + { + "epoch": 1.0436811374685462, + "grad_norm": 0.0011832509189844131, + "learning_rate": 0.13402004073824098, + "loss": 0.0882, + "num_input_tokens_seen": 27695776, + "step": 21360 + }, + { + "epoch": 1.0439254391322406, + "grad_norm": 0.0019990524742752314, + "learning_rate": 0.13396147232776062, + "loss": 0.1039, + "num_input_tokens_seen": 27702048, + "step": 21365 + }, + { + "epoch": 1.0441697407959347, + "grad_norm": 0.0013667969033122063, + "learning_rate": 0.13390290639062288, + "loss": 0.1027, + "num_input_tokens_seen": 27708640, + "step": 21370 + }, + { + "epoch": 1.044414042459629, + "grad_norm": 0.0018607144011184573, + "learning_rate": 0.13384434293585917, + "loss": 0.1075, + "num_input_tokens_seen": 27714496, + "step": 21375 + }, + { + "epoch": 1.0446583441233235, + "grad_norm": 0.0012680654181167483, + "learning_rate": 0.13378578197250088, + "loss": 0.0814, + "num_input_tokens_seen": 27721024, + "step": 21380 + }, + { + "epoch": 1.0449026457870179, + "grad_norm": 0.0010573461186140776, + "learning_rate": 0.13372722350957872, + "loss": 0.1008, + "num_input_tokens_seen": 27727584, + "step": 21385 + }, + { + "epoch": 1.0451469474507122, + "grad_norm": 0.0012509688967838883, + "learning_rate": 0.13366866755612322, + "loss": 0.0911, + "num_input_tokens_seen": 27733760, + "step": 21390 + }, + { + "epoch": 1.0453912491144064, + "grad_norm": 0.0014487792504951358, + "learning_rate": 0.13361011412116436, + "loss": 0.0812, + "num_input_tokens_seen": 27740192, + "step": 21395 + }, + { + "epoch": 1.0456355507781008, + "grad_norm": 0.0017140095587819815, + "learning_rate": 0.13355156321373196, + "loss": 0.1263, + "num_input_tokens_seen": 27746528, + "step": 21400 + }, + { + "epoch": 1.0456355507781008, + "eval_loss": 0.1003485918045044, + "eval_runtime": 402.5497, + "eval_samples_per_second": 90.386, + "eval_steps_per_second": 22.598, + "num_input_tokens_seen": 27746528, + "step": 21400 + }, + { + "epoch": 1.0458798524417952, + "grad_norm": 0.0012461612932384014, + "learning_rate": 0.13349301484285514, + "loss": 0.0792, + "num_input_tokens_seen": 27753216, + "step": 21405 + }, + { + "epoch": 1.0461241541054895, + "grad_norm": 0.002200531540438533, + "learning_rate": 0.13343446901756295, + "loss": 0.1051, + "num_input_tokens_seen": 27759936, + "step": 21410 + }, + { + "epoch": 1.0463684557691837, + "grad_norm": 0.002068073023110628, + "learning_rate": 0.13337592574688376, + "loss": 0.0928, + "num_input_tokens_seen": 27766400, + "step": 21415 + }, + { + "epoch": 1.046612757432878, + "grad_norm": 0.0009602477075532079, + "learning_rate": 0.13331738503984572, + "loss": 0.0974, + "num_input_tokens_seen": 27772704, + "step": 21420 + }, + { + "epoch": 1.0468570590965725, + "grad_norm": 0.001468145870603621, + "learning_rate": 0.1332588469054766, + "loss": 0.1092, + "num_input_tokens_seen": 27779296, + "step": 21425 + }, + { + "epoch": 1.0471013607602668, + "grad_norm": 0.001182141830213368, + "learning_rate": 0.1332003113528036, + "loss": 0.0859, + "num_input_tokens_seen": 27785408, + "step": 21430 + }, + { + "epoch": 1.0473456624239612, + "grad_norm": 0.0011585158063098788, + "learning_rate": 0.13314177839085373, + "loss": 0.1067, + "num_input_tokens_seen": 27791520, + "step": 21435 + }, + { + "epoch": 1.0475899640876554, + "grad_norm": 0.001268459134735167, + "learning_rate": 0.13308324802865354, + "loss": 0.0995, + "num_input_tokens_seen": 27798144, + "step": 21440 + }, + { + "epoch": 1.0478342657513497, + "grad_norm": 0.0007727954653091729, + "learning_rate": 0.13302472027522905, + "loss": 0.0756, + "num_input_tokens_seen": 27804768, + "step": 21445 + }, + { + "epoch": 1.0480785674150441, + "grad_norm": 0.0016053264262154698, + "learning_rate": 0.13296619513960606, + "loss": 0.1112, + "num_input_tokens_seen": 27811264, + "step": 21450 + }, + { + "epoch": 1.0483228690787385, + "grad_norm": 0.0010500488569959998, + "learning_rate": 0.1329076726308098, + "loss": 0.0904, + "num_input_tokens_seen": 27817664, + "step": 21455 + }, + { + "epoch": 1.0485671707424327, + "grad_norm": 0.0008867051801644266, + "learning_rate": 0.13284915275786519, + "loss": 0.0837, + "num_input_tokens_seen": 27824384, + "step": 21460 + }, + { + "epoch": 1.048811472406127, + "grad_norm": 0.0022801384329795837, + "learning_rate": 0.1327906355297968, + "loss": 0.0773, + "num_input_tokens_seen": 27830848, + "step": 21465 + }, + { + "epoch": 1.0490557740698214, + "grad_norm": 0.0009469336946494877, + "learning_rate": 0.13273212095562867, + "loss": 0.0874, + "num_input_tokens_seen": 27837568, + "step": 21470 + }, + { + "epoch": 1.0493000757335158, + "grad_norm": 0.0015141450567170978, + "learning_rate": 0.13267360904438444, + "loss": 0.082, + "num_input_tokens_seen": 27844224, + "step": 21475 + }, + { + "epoch": 1.0495443773972102, + "grad_norm": 0.0014495783252641559, + "learning_rate": 0.1326150998050875, + "loss": 0.0854, + "num_input_tokens_seen": 27850592, + "step": 21480 + }, + { + "epoch": 1.0497886790609043, + "grad_norm": 0.0007853690767660737, + "learning_rate": 0.1325565932467606, + "loss": 0.0794, + "num_input_tokens_seen": 27857376, + "step": 21485 + }, + { + "epoch": 1.0500329807245987, + "grad_norm": 0.0022418752778321505, + "learning_rate": 0.13249808937842628, + "loss": 0.146, + "num_input_tokens_seen": 27863872, + "step": 21490 + }, + { + "epoch": 1.050277282388293, + "grad_norm": 0.0015540429158136249, + "learning_rate": 0.1324395882091065, + "loss": 0.0972, + "num_input_tokens_seen": 27870272, + "step": 21495 + }, + { + "epoch": 1.0505215840519875, + "grad_norm": 0.0010854712454602122, + "learning_rate": 0.13238108974782284, + "loss": 0.0789, + "num_input_tokens_seen": 27876800, + "step": 21500 + }, + { + "epoch": 1.0507658857156816, + "grad_norm": 0.0010300107533112168, + "learning_rate": 0.13232259400359664, + "loss": 0.0786, + "num_input_tokens_seen": 27883584, + "step": 21505 + }, + { + "epoch": 1.051010187379376, + "grad_norm": 0.0008187037310563028, + "learning_rate": 0.13226410098544852, + "loss": 0.0882, + "num_input_tokens_seen": 27891072, + "step": 21510 + }, + { + "epoch": 1.0512544890430704, + "grad_norm": 0.0011580969439819455, + "learning_rate": 0.13220561070239892, + "loss": 0.0835, + "num_input_tokens_seen": 27898080, + "step": 21515 + }, + { + "epoch": 1.0514987907067648, + "grad_norm": 0.0009675098117440939, + "learning_rate": 0.13214712316346783, + "loss": 0.0781, + "num_input_tokens_seen": 27904768, + "step": 21520 + }, + { + "epoch": 1.051743092370459, + "grad_norm": 0.001962281297892332, + "learning_rate": 0.13208863837767465, + "loss": 0.1188, + "num_input_tokens_seen": 27911104, + "step": 21525 + }, + { + "epoch": 1.0519873940341533, + "grad_norm": 0.0013633311027660966, + "learning_rate": 0.13203015635403856, + "loss": 0.0895, + "num_input_tokens_seen": 27917376, + "step": 21530 + }, + { + "epoch": 1.0522316956978477, + "grad_norm": 0.0008904458954930305, + "learning_rate": 0.13197167710157817, + "loss": 0.0994, + "num_input_tokens_seen": 27924064, + "step": 21535 + }, + { + "epoch": 1.052475997361542, + "grad_norm": 0.000936213182285428, + "learning_rate": 0.13191320062931167, + "loss": 0.1048, + "num_input_tokens_seen": 27931264, + "step": 21540 + }, + { + "epoch": 1.0527202990252365, + "grad_norm": 0.0010667407186701894, + "learning_rate": 0.13185472694625702, + "loss": 0.1019, + "num_input_tokens_seen": 27937568, + "step": 21545 + }, + { + "epoch": 1.0529646006889306, + "grad_norm": 0.0013269615592435002, + "learning_rate": 0.13179625606143142, + "loss": 0.0967, + "num_input_tokens_seen": 27944128, + "step": 21550 + }, + { + "epoch": 1.053208902352625, + "grad_norm": 0.0010310201905667782, + "learning_rate": 0.13173778798385188, + "loss": 0.098, + "num_input_tokens_seen": 27950560, + "step": 21555 + }, + { + "epoch": 1.0534532040163194, + "grad_norm": 0.000711916945874691, + "learning_rate": 0.13167932272253505, + "loss": 0.082, + "num_input_tokens_seen": 27957152, + "step": 21560 + }, + { + "epoch": 1.0536975056800137, + "grad_norm": 0.0015295769553631544, + "learning_rate": 0.1316208602864968, + "loss": 0.1283, + "num_input_tokens_seen": 27963776, + "step": 21565 + }, + { + "epoch": 1.053941807343708, + "grad_norm": 0.0010906050447374582, + "learning_rate": 0.13156240068475292, + "loss": 0.1163, + "num_input_tokens_seen": 27970272, + "step": 21570 + }, + { + "epoch": 1.0541861090074023, + "grad_norm": 0.0008450300083495677, + "learning_rate": 0.1315039439263185, + "loss": 0.0818, + "num_input_tokens_seen": 27976544, + "step": 21575 + }, + { + "epoch": 1.0544304106710967, + "grad_norm": 0.0004230899503454566, + "learning_rate": 0.13144549002020833, + "loss": 0.0818, + "num_input_tokens_seen": 27982688, + "step": 21580 + }, + { + "epoch": 1.054674712334791, + "grad_norm": 0.0010178172960877419, + "learning_rate": 0.13138703897543688, + "loss": 0.0841, + "num_input_tokens_seen": 27989216, + "step": 21585 + }, + { + "epoch": 1.0549190139984854, + "grad_norm": 0.0014426193665713072, + "learning_rate": 0.1313285908010178, + "loss": 0.07, + "num_input_tokens_seen": 27996032, + "step": 21590 + }, + { + "epoch": 1.0551633156621796, + "grad_norm": 0.001193493022583425, + "learning_rate": 0.13127014550596475, + "loss": 0.1157, + "num_input_tokens_seen": 28002528, + "step": 21595 + }, + { + "epoch": 1.055407617325874, + "grad_norm": 0.0008787908591330051, + "learning_rate": 0.1312117030992906, + "loss": 0.0887, + "num_input_tokens_seen": 28009568, + "step": 21600 + }, + { + "epoch": 1.055407617325874, + "eval_loss": 0.0971703976392746, + "eval_runtime": 402.2664, + "eval_samples_per_second": 90.45, + "eval_steps_per_second": 22.614, + "num_input_tokens_seen": 28009568, + "step": 21600 + }, + { + "epoch": 1.0556519189895683, + "grad_norm": 0.0006513847038149834, + "learning_rate": 0.13115326359000795, + "loss": 0.0783, + "num_input_tokens_seen": 28016320, + "step": 21605 + }, + { + "epoch": 1.0558962206532627, + "grad_norm": 0.0011034030467271805, + "learning_rate": 0.13109482698712896, + "loss": 0.073, + "num_input_tokens_seen": 28023168, + "step": 21610 + }, + { + "epoch": 1.0561405223169569, + "grad_norm": 0.0004804357304237783, + "learning_rate": 0.1310363932996651, + "loss": 0.0546, + "num_input_tokens_seen": 28029408, + "step": 21615 + }, + { + "epoch": 1.0563848239806513, + "grad_norm": 0.0011482008267194033, + "learning_rate": 0.13097796253662775, + "loss": 0.0783, + "num_input_tokens_seen": 28035744, + "step": 21620 + }, + { + "epoch": 1.0566291256443456, + "grad_norm": 0.001988503150641918, + "learning_rate": 0.1309195347070277, + "loss": 0.08, + "num_input_tokens_seen": 28042784, + "step": 21625 + }, + { + "epoch": 1.05687342730804, + "grad_norm": 0.0009365826845169067, + "learning_rate": 0.13086110981987506, + "loss": 0.1173, + "num_input_tokens_seen": 28049504, + "step": 21630 + }, + { + "epoch": 1.0571177289717344, + "grad_norm": 0.0012571495026350021, + "learning_rate": 0.13080268788417987, + "loss": 0.1076, + "num_input_tokens_seen": 28055936, + "step": 21635 + }, + { + "epoch": 1.0573620306354286, + "grad_norm": 0.0004956864286214113, + "learning_rate": 0.1307442689089515, + "loss": 0.0925, + "num_input_tokens_seen": 28062496, + "step": 21640 + }, + { + "epoch": 1.057606332299123, + "grad_norm": 0.0015711060259491205, + "learning_rate": 0.13068585290319873, + "loss": 0.0857, + "num_input_tokens_seen": 28069632, + "step": 21645 + }, + { + "epoch": 1.0578506339628173, + "grad_norm": 0.0008950949413701892, + "learning_rate": 0.13062743987593026, + "loss": 0.0991, + "num_input_tokens_seen": 28075552, + "step": 21650 + }, + { + "epoch": 1.0580949356265117, + "grad_norm": 0.0010968934511765838, + "learning_rate": 0.13056902983615395, + "loss": 0.1014, + "num_input_tokens_seen": 28082016, + "step": 21655 + }, + { + "epoch": 1.0583392372902058, + "grad_norm": 0.0007000657496973872, + "learning_rate": 0.13051062279287742, + "loss": 0.0893, + "num_input_tokens_seen": 28088704, + "step": 21660 + }, + { + "epoch": 1.0585835389539002, + "grad_norm": 0.0008407121640630066, + "learning_rate": 0.13045221875510782, + "loss": 0.0835, + "num_input_tokens_seen": 28094752, + "step": 21665 + }, + { + "epoch": 1.0588278406175946, + "grad_norm": 0.0014512910274788737, + "learning_rate": 0.13039381773185174, + "loss": 0.0813, + "num_input_tokens_seen": 28101312, + "step": 21670 + }, + { + "epoch": 1.059072142281289, + "grad_norm": 0.0013088646810501814, + "learning_rate": 0.1303354197321153, + "loss": 0.0837, + "num_input_tokens_seen": 28107744, + "step": 21675 + }, + { + "epoch": 1.0593164439449834, + "grad_norm": 0.0014657687861472368, + "learning_rate": 0.13027702476490433, + "loss": 0.1049, + "num_input_tokens_seen": 28114368, + "step": 21680 + }, + { + "epoch": 1.0595607456086775, + "grad_norm": 0.0010215514339506626, + "learning_rate": 0.1302186328392239, + "loss": 0.0682, + "num_input_tokens_seen": 28120640, + "step": 21685 + }, + { + "epoch": 1.059805047272372, + "grad_norm": 0.0008938196115195751, + "learning_rate": 0.130160243964079, + "loss": 0.0762, + "num_input_tokens_seen": 28127232, + "step": 21690 + }, + { + "epoch": 1.0600493489360663, + "grad_norm": 0.0012243249220773578, + "learning_rate": 0.13010185814847372, + "loss": 0.0748, + "num_input_tokens_seen": 28133664, + "step": 21695 + }, + { + "epoch": 1.0602936505997607, + "grad_norm": 0.0008235318819060922, + "learning_rate": 0.13004347540141192, + "loss": 0.12, + "num_input_tokens_seen": 28140576, + "step": 21700 + }, + { + "epoch": 1.0605379522634548, + "grad_norm": 0.0010288581252098083, + "learning_rate": 0.12998509573189712, + "loss": 0.079, + "num_input_tokens_seen": 28146848, + "step": 21705 + }, + { + "epoch": 1.0607822539271492, + "grad_norm": 0.0013855879660695791, + "learning_rate": 0.12992671914893203, + "loss": 0.1129, + "num_input_tokens_seen": 28153152, + "step": 21710 + }, + { + "epoch": 1.0610265555908436, + "grad_norm": 0.001112298690713942, + "learning_rate": 0.12986834566151909, + "loss": 0.0597, + "num_input_tokens_seen": 28160000, + "step": 21715 + }, + { + "epoch": 1.061270857254538, + "grad_norm": 0.001267043873667717, + "learning_rate": 0.12980997527866028, + "loss": 0.104, + "num_input_tokens_seen": 28166944, + "step": 21720 + }, + { + "epoch": 1.0615151589182323, + "grad_norm": 0.0018657655455172062, + "learning_rate": 0.12975160800935692, + "loss": 0.0816, + "num_input_tokens_seen": 28173824, + "step": 21725 + }, + { + "epoch": 1.0617594605819265, + "grad_norm": 0.001341348048299551, + "learning_rate": 0.12969324386261016, + "loss": 0.1029, + "num_input_tokens_seen": 28180064, + "step": 21730 + }, + { + "epoch": 1.0620037622456209, + "grad_norm": 0.00087919388897717, + "learning_rate": 0.12963488284742034, + "loss": 0.1042, + "num_input_tokens_seen": 28186816, + "step": 21735 + }, + { + "epoch": 1.0622480639093153, + "grad_norm": 0.0008439011289738119, + "learning_rate": 0.12957652497278752, + "loss": 0.0713, + "num_input_tokens_seen": 28193440, + "step": 21740 + }, + { + "epoch": 1.0624923655730096, + "grad_norm": 0.0010903330985456705, + "learning_rate": 0.12951817024771117, + "loss": 0.0662, + "num_input_tokens_seen": 28200032, + "step": 21745 + }, + { + "epoch": 1.0627366672367038, + "grad_norm": 0.000914389209356159, + "learning_rate": 0.12945981868119041, + "loss": 0.1024, + "num_input_tokens_seen": 28206304, + "step": 21750 + }, + { + "epoch": 1.0629809689003982, + "grad_norm": 0.0010625508148223162, + "learning_rate": 0.12940147028222376, + "loss": 0.079, + "num_input_tokens_seen": 28212800, + "step": 21755 + }, + { + "epoch": 1.0632252705640926, + "grad_norm": 0.0008063875138759613, + "learning_rate": 0.12934312505980916, + "loss": 0.091, + "num_input_tokens_seen": 28219200, + "step": 21760 + }, + { + "epoch": 1.063469572227787, + "grad_norm": 0.0006029783398844302, + "learning_rate": 0.1292847830229443, + "loss": 0.0807, + "num_input_tokens_seen": 28225472, + "step": 21765 + }, + { + "epoch": 1.0637138738914813, + "grad_norm": 0.0016971477307379246, + "learning_rate": 0.12922644418062626, + "loss": 0.0795, + "num_input_tokens_seen": 28232064, + "step": 21770 + }, + { + "epoch": 1.0639581755551755, + "grad_norm": 0.0009886592160910368, + "learning_rate": 0.1291681085418515, + "loss": 0.0897, + "num_input_tokens_seen": 28238528, + "step": 21775 + }, + { + "epoch": 1.0642024772188698, + "grad_norm": 0.0012404738226905465, + "learning_rate": 0.12910977611561628, + "loss": 0.1163, + "num_input_tokens_seen": 28245120, + "step": 21780 + }, + { + "epoch": 1.0644467788825642, + "grad_norm": 0.0010038241744041443, + "learning_rate": 0.1290514469109161, + "loss": 0.1209, + "num_input_tokens_seen": 28251328, + "step": 21785 + }, + { + "epoch": 1.0646910805462586, + "grad_norm": 0.0018177377060055733, + "learning_rate": 0.128993120936746, + "loss": 0.0794, + "num_input_tokens_seen": 28257696, + "step": 21790 + }, + { + "epoch": 1.0649353822099528, + "grad_norm": 0.0016259591793641448, + "learning_rate": 0.12893479820210071, + "loss": 0.0955, + "num_input_tokens_seen": 28264256, + "step": 21795 + }, + { + "epoch": 1.0651796838736471, + "grad_norm": 0.0016881373012438416, + "learning_rate": 0.1288764787159742, + "loss": 0.0905, + "num_input_tokens_seen": 28270592, + "step": 21800 + }, + { + "epoch": 1.0651796838736471, + "eval_loss": 0.09476463496685028, + "eval_runtime": 402.7542, + "eval_samples_per_second": 90.34, + "eval_steps_per_second": 22.587, + "num_input_tokens_seen": 28270592, + "step": 21800 + }, + { + "epoch": 1.0654239855373415, + "grad_norm": 0.0013624971034005284, + "learning_rate": 0.1288181624873601, + "loss": 0.1018, + "num_input_tokens_seen": 28277632, + "step": 21805 + }, + { + "epoch": 1.065668287201036, + "grad_norm": 0.001085534691810608, + "learning_rate": 0.12875984952525163, + "loss": 0.0625, + "num_input_tokens_seen": 28283872, + "step": 21810 + }, + { + "epoch": 1.0659125888647303, + "grad_norm": 0.0014808244304731488, + "learning_rate": 0.12870153983864122, + "loss": 0.1044, + "num_input_tokens_seen": 28290112, + "step": 21815 + }, + { + "epoch": 1.0661568905284244, + "grad_norm": 0.0008906896691769361, + "learning_rate": 0.12864323343652104, + "loss": 0.0967, + "num_input_tokens_seen": 28296640, + "step": 21820 + }, + { + "epoch": 1.0664011921921188, + "grad_norm": 0.0008551803766749799, + "learning_rate": 0.12858493032788268, + "loss": 0.1087, + "num_input_tokens_seen": 28303520, + "step": 21825 + }, + { + "epoch": 1.0666454938558132, + "grad_norm": 0.0023533375933766365, + "learning_rate": 0.12852663052171714, + "loss": 0.1018, + "num_input_tokens_seen": 28309824, + "step": 21830 + }, + { + "epoch": 1.0668897955195076, + "grad_norm": 0.0010609121527522802, + "learning_rate": 0.12846833402701507, + "loss": 0.0911, + "num_input_tokens_seen": 28316256, + "step": 21835 + }, + { + "epoch": 1.0671340971832017, + "grad_norm": 0.0014248040970414877, + "learning_rate": 0.12841004085276642, + "loss": 0.0728, + "num_input_tokens_seen": 28323040, + "step": 21840 + }, + { + "epoch": 1.0673783988468961, + "grad_norm": 0.0008048378513194621, + "learning_rate": 0.12835175100796076, + "loss": 0.0762, + "num_input_tokens_seen": 28329824, + "step": 21845 + }, + { + "epoch": 1.0676227005105905, + "grad_norm": 0.0013174728956073523, + "learning_rate": 0.12829346450158724, + "loss": 0.0856, + "num_input_tokens_seen": 28335968, + "step": 21850 + }, + { + "epoch": 1.0678670021742849, + "grad_norm": 0.0009695563348941505, + "learning_rate": 0.12823518134263423, + "loss": 0.0962, + "num_input_tokens_seen": 28343008, + "step": 21855 + }, + { + "epoch": 1.0681113038379793, + "grad_norm": 0.0013702824944630265, + "learning_rate": 0.12817690154008973, + "loss": 0.0902, + "num_input_tokens_seen": 28349760, + "step": 21860 + }, + { + "epoch": 1.0683556055016734, + "grad_norm": 0.001774936099536717, + "learning_rate": 0.12811862510294134, + "loss": 0.081, + "num_input_tokens_seen": 28355840, + "step": 21865 + }, + { + "epoch": 1.0685999071653678, + "grad_norm": 0.0007971441955305636, + "learning_rate": 0.12806035204017585, + "loss": 0.0807, + "num_input_tokens_seen": 28362016, + "step": 21870 + }, + { + "epoch": 1.0688442088290622, + "grad_norm": 0.0020528153982013464, + "learning_rate": 0.12800208236077987, + "loss": 0.0925, + "num_input_tokens_seen": 28368512, + "step": 21875 + }, + { + "epoch": 1.0690885104927565, + "grad_norm": 0.0016294163651764393, + "learning_rate": 0.12794381607373917, + "loss": 0.078, + "num_input_tokens_seen": 28374976, + "step": 21880 + }, + { + "epoch": 1.0693328121564507, + "grad_norm": 0.001085562282241881, + "learning_rate": 0.12788555318803924, + "loss": 0.1197, + "num_input_tokens_seen": 28381792, + "step": 21885 + }, + { + "epoch": 1.069577113820145, + "grad_norm": 0.0010524754179641604, + "learning_rate": 0.1278272937126649, + "loss": 0.1126, + "num_input_tokens_seen": 28388128, + "step": 21890 + }, + { + "epoch": 1.0698214154838395, + "grad_norm": 0.0013323462335392833, + "learning_rate": 0.1277690376566005, + "loss": 0.0942, + "num_input_tokens_seen": 28394720, + "step": 21895 + }, + { + "epoch": 1.0700657171475338, + "grad_norm": 0.0014220719458535314, + "learning_rate": 0.12771078502882985, + "loss": 0.0907, + "num_input_tokens_seen": 28401600, + "step": 21900 + }, + { + "epoch": 1.070310018811228, + "grad_norm": 0.0012859145645052195, + "learning_rate": 0.12765253583833633, + "loss": 0.0809, + "num_input_tokens_seen": 28408160, + "step": 21905 + }, + { + "epoch": 1.0705543204749224, + "grad_norm": 0.0010528876446187496, + "learning_rate": 0.12759429009410256, + "loss": 0.0983, + "num_input_tokens_seen": 28414528, + "step": 21910 + }, + { + "epoch": 1.0707986221386168, + "grad_norm": 0.0010539523791521788, + "learning_rate": 0.12753604780511085, + "loss": 0.1034, + "num_input_tokens_seen": 28420864, + "step": 21915 + }, + { + "epoch": 1.0710429238023111, + "grad_norm": 0.0009950114181265235, + "learning_rate": 0.12747780898034283, + "loss": 0.1007, + "num_input_tokens_seen": 28427776, + "step": 21920 + }, + { + "epoch": 1.0712872254660055, + "grad_norm": 0.0013359255390241742, + "learning_rate": 0.12741957362877973, + "loss": 0.0956, + "num_input_tokens_seen": 28434176, + "step": 21925 + }, + { + "epoch": 1.0715315271296997, + "grad_norm": 0.0011911304900422692, + "learning_rate": 0.12736134175940214, + "loss": 0.1019, + "num_input_tokens_seen": 28440896, + "step": 21930 + }, + { + "epoch": 1.071775828793394, + "grad_norm": 0.0013209469616413116, + "learning_rate": 0.12730311338119016, + "loss": 0.1008, + "num_input_tokens_seen": 28447584, + "step": 21935 + }, + { + "epoch": 1.0720201304570884, + "grad_norm": 0.0007417346350848675, + "learning_rate": 0.12724488850312327, + "loss": 0.0889, + "num_input_tokens_seen": 28454016, + "step": 21940 + }, + { + "epoch": 1.0722644321207828, + "grad_norm": 0.0010617448715493083, + "learning_rate": 0.1271866671341806, + "loss": 0.0982, + "num_input_tokens_seen": 28460736, + "step": 21945 + }, + { + "epoch": 1.072508733784477, + "grad_norm": 0.0005201597814448178, + "learning_rate": 0.12712844928334047, + "loss": 0.0835, + "num_input_tokens_seen": 28467424, + "step": 21950 + }, + { + "epoch": 1.0727530354481714, + "grad_norm": 0.0017951990012079477, + "learning_rate": 0.12707023495958095, + "loss": 0.1019, + "num_input_tokens_seen": 28473920, + "step": 21955 + }, + { + "epoch": 1.0729973371118657, + "grad_norm": 0.0009675905457697809, + "learning_rate": 0.12701202417187932, + "loss": 0.0806, + "num_input_tokens_seen": 28480576, + "step": 21960 + }, + { + "epoch": 1.0732416387755601, + "grad_norm": 0.0010810394305735826, + "learning_rate": 0.12695381692921243, + "loss": 0.0998, + "num_input_tokens_seen": 28486720, + "step": 21965 + }, + { + "epoch": 1.0734859404392545, + "grad_norm": 0.001334147178567946, + "learning_rate": 0.12689561324055665, + "loss": 0.0993, + "num_input_tokens_seen": 28493696, + "step": 21970 + }, + { + "epoch": 1.0737302421029487, + "grad_norm": 0.001329729682765901, + "learning_rate": 0.12683741311488758, + "loss": 0.1031, + "num_input_tokens_seen": 28499968, + "step": 21975 + }, + { + "epoch": 1.073974543766643, + "grad_norm": 0.0018884571036323905, + "learning_rate": 0.1267792165611805, + "loss": 0.0731, + "num_input_tokens_seen": 28506976, + "step": 21980 + }, + { + "epoch": 1.0742188454303374, + "grad_norm": 0.001368609257042408, + "learning_rate": 0.1267210235884101, + "loss": 0.073, + "num_input_tokens_seen": 28513792, + "step": 21985 + }, + { + "epoch": 1.0744631470940318, + "grad_norm": 0.001095268758945167, + "learning_rate": 0.12666283420555033, + "loss": 0.084, + "num_input_tokens_seen": 28520608, + "step": 21990 + }, + { + "epoch": 1.074707448757726, + "grad_norm": 0.0008243289776146412, + "learning_rate": 0.12660464842157487, + "loss": 0.1045, + "num_input_tokens_seen": 28526944, + "step": 21995 + }, + { + "epoch": 1.0749517504214203, + "grad_norm": 0.0010260287672281265, + "learning_rate": 0.1265464662454566, + "loss": 0.0639, + "num_input_tokens_seen": 28533952, + "step": 22000 + }, + { + "epoch": 1.0749517504214203, + "eval_loss": 0.09810065478086472, + "eval_runtime": 402.8647, + "eval_samples_per_second": 90.316, + "eval_steps_per_second": 22.581, + "num_input_tokens_seen": 28533952, + "step": 22000 + }, + { + "epoch": 1.0751960520851147, + "grad_norm": 0.001172712305560708, + "learning_rate": 0.12648828768616793, + "loss": 0.1006, + "num_input_tokens_seen": 28540576, + "step": 22005 + }, + { + "epoch": 1.075440353748809, + "grad_norm": 0.0012334393104538321, + "learning_rate": 0.12643011275268085, + "loss": 0.1165, + "num_input_tokens_seen": 28547264, + "step": 22010 + }, + { + "epoch": 1.0756846554125032, + "grad_norm": 0.0010112286545336246, + "learning_rate": 0.1263719414539665, + "loss": 0.0791, + "num_input_tokens_seen": 28553120, + "step": 22015 + }, + { + "epoch": 1.0759289570761976, + "grad_norm": 0.001184401335194707, + "learning_rate": 0.1263137737989957, + "loss": 0.1269, + "num_input_tokens_seen": 28559104, + "step": 22020 + }, + { + "epoch": 1.076173258739892, + "grad_norm": 0.0011576494434848428, + "learning_rate": 0.1262556097967387, + "loss": 0.0876, + "num_input_tokens_seen": 28565440, + "step": 22025 + }, + { + "epoch": 1.0764175604035864, + "grad_norm": 0.0008827525889500976, + "learning_rate": 0.126197449456165, + "loss": 0.096, + "num_input_tokens_seen": 28572000, + "step": 22030 + }, + { + "epoch": 1.0766618620672808, + "grad_norm": 0.0014898793306201696, + "learning_rate": 0.12613929278624378, + "loss": 0.1071, + "num_input_tokens_seen": 28578784, + "step": 22035 + }, + { + "epoch": 1.076906163730975, + "grad_norm": 0.00125587556976825, + "learning_rate": 0.12608113979594343, + "loss": 0.1019, + "num_input_tokens_seen": 28585120, + "step": 22040 + }, + { + "epoch": 1.0771504653946693, + "grad_norm": 0.0006617777980864048, + "learning_rate": 0.1260229904942319, + "loss": 0.0702, + "num_input_tokens_seen": 28591584, + "step": 22045 + }, + { + "epoch": 1.0773947670583637, + "grad_norm": 0.0010538690257817507, + "learning_rate": 0.12596484489007662, + "loss": 0.103, + "num_input_tokens_seen": 28597792, + "step": 22050 + }, + { + "epoch": 1.077639068722058, + "grad_norm": 0.0009558504680171609, + "learning_rate": 0.1259067029924442, + "loss": 0.0963, + "num_input_tokens_seen": 28604064, + "step": 22055 + }, + { + "epoch": 1.0778833703857522, + "grad_norm": 0.001322999014519155, + "learning_rate": 0.12584856481030096, + "loss": 0.0989, + "num_input_tokens_seen": 28610336, + "step": 22060 + }, + { + "epoch": 1.0781276720494466, + "grad_norm": 0.002006783150136471, + "learning_rate": 0.12579043035261261, + "loss": 0.0858, + "num_input_tokens_seen": 28617344, + "step": 22065 + }, + { + "epoch": 1.078371973713141, + "grad_norm": 0.0019091736758127809, + "learning_rate": 0.1257322996283441, + "loss": 0.1075, + "num_input_tokens_seen": 28623296, + "step": 22070 + }, + { + "epoch": 1.0786162753768354, + "grad_norm": 0.0010439192410558462, + "learning_rate": 0.12567417264645994, + "loss": 0.0633, + "num_input_tokens_seen": 28629952, + "step": 22075 + }, + { + "epoch": 1.0788605770405297, + "grad_norm": 0.0007738082204014063, + "learning_rate": 0.12561604941592408, + "loss": 0.113, + "num_input_tokens_seen": 28636192, + "step": 22080 + }, + { + "epoch": 1.079104878704224, + "grad_norm": 0.0019593180622905493, + "learning_rate": 0.12555792994569978, + "loss": 0.0956, + "num_input_tokens_seen": 28642816, + "step": 22085 + }, + { + "epoch": 1.0793491803679183, + "grad_norm": 0.001536033465526998, + "learning_rate": 0.1254998142447499, + "loss": 0.0917, + "num_input_tokens_seen": 28648672, + "step": 22090 + }, + { + "epoch": 1.0795934820316127, + "grad_norm": 0.0007655537920072675, + "learning_rate": 0.1254417023220365, + "loss": 0.0959, + "num_input_tokens_seen": 28654784, + "step": 22095 + }, + { + "epoch": 1.079837783695307, + "grad_norm": 0.00149181904271245, + "learning_rate": 0.12538359418652126, + "loss": 0.088, + "num_input_tokens_seen": 28661376, + "step": 22100 + }, + { + "epoch": 1.0800820853590012, + "grad_norm": 0.0011864706175401807, + "learning_rate": 0.12532548984716513, + "loss": 0.0923, + "num_input_tokens_seen": 28667712, + "step": 22105 + }, + { + "epoch": 1.0803263870226956, + "grad_norm": 0.000873398152180016, + "learning_rate": 0.12526738931292855, + "loss": 0.0855, + "num_input_tokens_seen": 28674208, + "step": 22110 + }, + { + "epoch": 1.08057068868639, + "grad_norm": 0.0011393956374377012, + "learning_rate": 0.1252092925927714, + "loss": 0.0827, + "num_input_tokens_seen": 28680736, + "step": 22115 + }, + { + "epoch": 1.0808149903500843, + "grad_norm": 0.0012329547898843884, + "learning_rate": 0.12515119969565278, + "loss": 0.0725, + "num_input_tokens_seen": 28686848, + "step": 22120 + }, + { + "epoch": 1.0810592920137787, + "grad_norm": 0.0016241817502304912, + "learning_rate": 0.12509311063053144, + "loss": 0.0998, + "num_input_tokens_seen": 28692768, + "step": 22125 + }, + { + "epoch": 1.0813035936774729, + "grad_norm": 0.0011284807696938515, + "learning_rate": 0.1250350254063655, + "loss": 0.0768, + "num_input_tokens_seen": 28699072, + "step": 22130 + }, + { + "epoch": 1.0815478953411672, + "grad_norm": 0.001408719806931913, + "learning_rate": 0.1249769440321123, + "loss": 0.0888, + "num_input_tokens_seen": 28705376, + "step": 22135 + }, + { + "epoch": 1.0817921970048616, + "grad_norm": 0.0014837068738415837, + "learning_rate": 0.12491886651672884, + "loss": 0.0977, + "num_input_tokens_seen": 28711808, + "step": 22140 + }, + { + "epoch": 1.082036498668556, + "grad_norm": 0.0020732558332383633, + "learning_rate": 0.12486079286917139, + "loss": 0.1005, + "num_input_tokens_seen": 28717920, + "step": 22145 + }, + { + "epoch": 1.0822808003322502, + "grad_norm": 0.0013993524480611086, + "learning_rate": 0.12480272309839553, + "loss": 0.0985, + "num_input_tokens_seen": 28724128, + "step": 22150 + }, + { + "epoch": 1.0825251019959445, + "grad_norm": 0.0017567953327670693, + "learning_rate": 0.12474465721335648, + "loss": 0.1147, + "num_input_tokens_seen": 28730304, + "step": 22155 + }, + { + "epoch": 1.082769403659639, + "grad_norm": 0.001232940354384482, + "learning_rate": 0.12468659522300861, + "loss": 0.1043, + "num_input_tokens_seen": 28737440, + "step": 22160 + }, + { + "epoch": 1.0830137053233333, + "grad_norm": 0.0013608860317617655, + "learning_rate": 0.12462853713630584, + "loss": 0.1127, + "num_input_tokens_seen": 28743840, + "step": 22165 + }, + { + "epoch": 1.0832580069870277, + "grad_norm": 0.000591309682931751, + "learning_rate": 0.12457048296220156, + "loss": 0.1076, + "num_input_tokens_seen": 28750048, + "step": 22170 + }, + { + "epoch": 1.0835023086507218, + "grad_norm": 0.0012762482510879636, + "learning_rate": 0.12451243270964832, + "loss": 0.0921, + "num_input_tokens_seen": 28756704, + "step": 22175 + }, + { + "epoch": 1.0837466103144162, + "grad_norm": 0.002171242842450738, + "learning_rate": 0.12445438638759827, + "loss": 0.1021, + "num_input_tokens_seen": 28762912, + "step": 22180 + }, + { + "epoch": 1.0839909119781106, + "grad_norm": 0.0013401230098679662, + "learning_rate": 0.1243963440050029, + "loss": 0.0876, + "num_input_tokens_seen": 28769152, + "step": 22185 + }, + { + "epoch": 1.084235213641805, + "grad_norm": 0.0012115460122004151, + "learning_rate": 0.12433830557081298, + "loss": 0.0546, + "num_input_tokens_seen": 28775744, + "step": 22190 + }, + { + "epoch": 1.0844795153054991, + "grad_norm": 0.001222984865307808, + "learning_rate": 0.12428027109397889, + "loss": 0.1077, + "num_input_tokens_seen": 28782080, + "step": 22195 + }, + { + "epoch": 1.0847238169691935, + "grad_norm": 0.0010363858891651034, + "learning_rate": 0.12422224058345015, + "loss": 0.092, + "num_input_tokens_seen": 28788352, + "step": 22200 + }, + { + "epoch": 1.0847238169691935, + "eval_loss": 0.09352832287549973, + "eval_runtime": 402.5702, + "eval_samples_per_second": 90.382, + "eval_steps_per_second": 22.597, + "num_input_tokens_seen": 28788352, + "step": 22200 + }, + { + "epoch": 1.084968118632888, + "grad_norm": 0.0009480526787228882, + "learning_rate": 0.12416421404817583, + "loss": 0.074, + "num_input_tokens_seen": 28795072, + "step": 22205 + }, + { + "epoch": 1.0852124202965823, + "grad_norm": 0.0007700546993874013, + "learning_rate": 0.12410619149710447, + "loss": 0.1075, + "num_input_tokens_seen": 28801824, + "step": 22210 + }, + { + "epoch": 1.0854567219602766, + "grad_norm": 0.0015195320593193173, + "learning_rate": 0.12404817293918374, + "loss": 0.105, + "num_input_tokens_seen": 28808480, + "step": 22215 + }, + { + "epoch": 1.0857010236239708, + "grad_norm": 0.002098476281389594, + "learning_rate": 0.12399015838336086, + "loss": 0.1099, + "num_input_tokens_seen": 28815328, + "step": 22220 + }, + { + "epoch": 1.0859453252876652, + "grad_norm": 0.0018052667146548629, + "learning_rate": 0.12393214783858246, + "loss": 0.1036, + "num_input_tokens_seen": 28821728, + "step": 22225 + }, + { + "epoch": 1.0861896269513596, + "grad_norm": 0.0038209580816328526, + "learning_rate": 0.1238741413137944, + "loss": 0.0961, + "num_input_tokens_seen": 28828192, + "step": 22230 + }, + { + "epoch": 1.086433928615054, + "grad_norm": 0.0013519618660211563, + "learning_rate": 0.12381613881794212, + "loss": 0.113, + "num_input_tokens_seen": 28834272, + "step": 22235 + }, + { + "epoch": 1.086678230278748, + "grad_norm": 0.0013146541314199567, + "learning_rate": 0.12375814035997022, + "loss": 0.096, + "num_input_tokens_seen": 28840384, + "step": 22240 + }, + { + "epoch": 1.0869225319424425, + "grad_norm": 0.001735515077598393, + "learning_rate": 0.12370014594882285, + "loss": 0.1077, + "num_input_tokens_seen": 28846816, + "step": 22245 + }, + { + "epoch": 1.0871668336061369, + "grad_norm": 0.0012084910413250327, + "learning_rate": 0.12364215559344356, + "loss": 0.0984, + "num_input_tokens_seen": 28853024, + "step": 22250 + }, + { + "epoch": 1.0874111352698312, + "grad_norm": 0.0015198498731479049, + "learning_rate": 0.12358416930277506, + "loss": 0.1117, + "num_input_tokens_seen": 28859552, + "step": 22255 + }, + { + "epoch": 1.0876554369335256, + "grad_norm": 0.0011666014324873686, + "learning_rate": 0.1235261870857596, + "loss": 0.1339, + "num_input_tokens_seen": 28866720, + "step": 22260 + }, + { + "epoch": 1.0878997385972198, + "grad_norm": 0.0008529219194315374, + "learning_rate": 0.12346820895133884, + "loss": 0.0772, + "num_input_tokens_seen": 28873280, + "step": 22265 + }, + { + "epoch": 1.0881440402609142, + "grad_norm": 0.001574606285430491, + "learning_rate": 0.12341023490845361, + "loss": 0.1046, + "num_input_tokens_seen": 28880128, + "step": 22270 + }, + { + "epoch": 1.0883883419246085, + "grad_norm": 0.0007719980203546584, + "learning_rate": 0.12335226496604437, + "loss": 0.0894, + "num_input_tokens_seen": 28886720, + "step": 22275 + }, + { + "epoch": 1.088632643588303, + "grad_norm": 0.0010042217327281833, + "learning_rate": 0.12329429913305069, + "loss": 0.0845, + "num_input_tokens_seen": 28893088, + "step": 22280 + }, + { + "epoch": 1.088876945251997, + "grad_norm": 0.0009649527492001653, + "learning_rate": 0.12323633741841171, + "loss": 0.0727, + "num_input_tokens_seen": 28899904, + "step": 22285 + }, + { + "epoch": 1.0891212469156915, + "grad_norm": 0.0013962194789201021, + "learning_rate": 0.12317837983106583, + "loss": 0.1063, + "num_input_tokens_seen": 28906176, + "step": 22290 + }, + { + "epoch": 1.0893655485793858, + "grad_norm": 0.0013276386307552457, + "learning_rate": 0.12312042637995087, + "loss": 0.0873, + "num_input_tokens_seen": 28912416, + "step": 22295 + }, + { + "epoch": 1.0896098502430802, + "grad_norm": 0.0019295443780720234, + "learning_rate": 0.12306247707400389, + "loss": 0.0911, + "num_input_tokens_seen": 28918976, + "step": 22300 + }, + { + "epoch": 1.0898541519067746, + "grad_norm": 0.0010823950869962573, + "learning_rate": 0.12300453192216154, + "loss": 0.0848, + "num_input_tokens_seen": 28925056, + "step": 22305 + }, + { + "epoch": 1.0900984535704688, + "grad_norm": 0.0018551347311586142, + "learning_rate": 0.12294659093335956, + "loss": 0.1081, + "num_input_tokens_seen": 28931520, + "step": 22310 + }, + { + "epoch": 1.0903427552341631, + "grad_norm": 0.0016293056542053819, + "learning_rate": 0.12288865411653327, + "loss": 0.0948, + "num_input_tokens_seen": 28937760, + "step": 22315 + }, + { + "epoch": 1.0905870568978575, + "grad_norm": 0.0010726905893534422, + "learning_rate": 0.12283072148061717, + "loss": 0.0972, + "num_input_tokens_seen": 28943872, + "step": 22320 + }, + { + "epoch": 1.090831358561552, + "grad_norm": 0.001135341008193791, + "learning_rate": 0.12277279303454529, + "loss": 0.1092, + "num_input_tokens_seen": 28950112, + "step": 22325 + }, + { + "epoch": 1.091075660225246, + "grad_norm": 0.0009680751827545464, + "learning_rate": 0.12271486878725091, + "loss": 0.0973, + "num_input_tokens_seen": 28956640, + "step": 22330 + }, + { + "epoch": 1.0913199618889404, + "grad_norm": 0.0012115162098780274, + "learning_rate": 0.12265694874766658, + "loss": 0.0827, + "num_input_tokens_seen": 28963136, + "step": 22335 + }, + { + "epoch": 1.0915642635526348, + "grad_norm": 0.0012400184059515595, + "learning_rate": 0.12259903292472435, + "loss": 0.0725, + "num_input_tokens_seen": 28969376, + "step": 22340 + }, + { + "epoch": 1.0918085652163292, + "grad_norm": 0.0011850749142467976, + "learning_rate": 0.12254112132735567, + "loss": 0.0777, + "num_input_tokens_seen": 28975968, + "step": 22345 + }, + { + "epoch": 1.0920528668800236, + "grad_norm": 0.0017058472149074078, + "learning_rate": 0.12248321396449108, + "loss": 0.0925, + "num_input_tokens_seen": 28982368, + "step": 22350 + }, + { + "epoch": 1.0922971685437177, + "grad_norm": 0.0009577056625857949, + "learning_rate": 0.12242531084506075, + "loss": 0.0747, + "num_input_tokens_seen": 28989216, + "step": 22355 + }, + { + "epoch": 1.092541470207412, + "grad_norm": 0.0008254505810327828, + "learning_rate": 0.122367411977994, + "loss": 0.068, + "num_input_tokens_seen": 28995872, + "step": 22360 + }, + { + "epoch": 1.0927857718711065, + "grad_norm": 0.0012112975819036365, + "learning_rate": 0.12230951737221954, + "loss": 0.0795, + "num_input_tokens_seen": 29002624, + "step": 22365 + }, + { + "epoch": 1.0930300735348009, + "grad_norm": 0.0018581878393888474, + "learning_rate": 0.12225162703666555, + "loss": 0.0819, + "num_input_tokens_seen": 29009184, + "step": 22370 + }, + { + "epoch": 1.093274375198495, + "grad_norm": 0.0009380063856951892, + "learning_rate": 0.1221937409802593, + "loss": 0.0981, + "num_input_tokens_seen": 29015648, + "step": 22375 + }, + { + "epoch": 1.0935186768621894, + "grad_norm": 0.0015639775665476918, + "learning_rate": 0.12213585921192768, + "loss": 0.0967, + "num_input_tokens_seen": 29022144, + "step": 22380 + }, + { + "epoch": 1.0937629785258838, + "grad_norm": 0.001674798084422946, + "learning_rate": 0.1220779817405967, + "loss": 0.0879, + "num_input_tokens_seen": 29028032, + "step": 22385 + }, + { + "epoch": 1.0940072801895782, + "grad_norm": 0.0016830076929181814, + "learning_rate": 0.12202010857519181, + "loss": 0.0995, + "num_input_tokens_seen": 29034720, + "step": 22390 + }, + { + "epoch": 1.0942515818532725, + "grad_norm": 0.0010319831781089306, + "learning_rate": 0.12196223972463785, + "loss": 0.0909, + "num_input_tokens_seen": 29040864, + "step": 22395 + }, + { + "epoch": 1.0944958835169667, + "grad_norm": 0.0010576698696240783, + "learning_rate": 0.12190437519785885, + "loss": 0.0877, + "num_input_tokens_seen": 29047328, + "step": 22400 + }, + { + "epoch": 1.0944958835169667, + "eval_loss": 0.09437805414199829, + "eval_runtime": 402.1394, + "eval_samples_per_second": 90.479, + "eval_steps_per_second": 22.622, + "num_input_tokens_seen": 29047328, + "step": 22400 + }, + { + "epoch": 1.094740185180661, + "grad_norm": 0.002498387824743986, + "learning_rate": 0.12184651500377823, + "loss": 0.1077, + "num_input_tokens_seen": 29053504, + "step": 22405 + }, + { + "epoch": 1.0949844868443555, + "grad_norm": 0.0011763314250856638, + "learning_rate": 0.12178865915131885, + "loss": 0.0982, + "num_input_tokens_seen": 29059872, + "step": 22410 + }, + { + "epoch": 1.0952287885080498, + "grad_norm": 0.0012885077157989144, + "learning_rate": 0.1217308076494027, + "loss": 0.0903, + "num_input_tokens_seen": 29066176, + "step": 22415 + }, + { + "epoch": 1.095473090171744, + "grad_norm": 0.001056781387887895, + "learning_rate": 0.12167296050695134, + "loss": 0.0744, + "num_input_tokens_seen": 29072480, + "step": 22420 + }, + { + "epoch": 1.0957173918354384, + "grad_norm": 0.0017068303423002362, + "learning_rate": 0.12161511773288536, + "loss": 0.108, + "num_input_tokens_seen": 29078880, + "step": 22425 + }, + { + "epoch": 1.0959616934991327, + "grad_norm": 0.0016942678485065699, + "learning_rate": 0.121557279336125, + "loss": 0.0907, + "num_input_tokens_seen": 29084928, + "step": 22430 + }, + { + "epoch": 1.0962059951628271, + "grad_norm": 0.0007431362755596638, + "learning_rate": 0.12149944532558957, + "loss": 0.102, + "num_input_tokens_seen": 29091072, + "step": 22435 + }, + { + "epoch": 1.0964502968265213, + "grad_norm": 0.001760152168571949, + "learning_rate": 0.12144161571019785, + "loss": 0.109, + "num_input_tokens_seen": 29097280, + "step": 22440 + }, + { + "epoch": 1.0966945984902157, + "grad_norm": 0.0008649717201478779, + "learning_rate": 0.12138379049886781, + "loss": 0.086, + "num_input_tokens_seen": 29103872, + "step": 22445 + }, + { + "epoch": 1.09693890015391, + "grad_norm": 0.0015086558414623141, + "learning_rate": 0.12132596970051697, + "loss": 0.0892, + "num_input_tokens_seen": 29110688, + "step": 22450 + }, + { + "epoch": 1.0971832018176044, + "grad_norm": 0.0015327486908063293, + "learning_rate": 0.12126815332406189, + "loss": 0.1182, + "num_input_tokens_seen": 29117376, + "step": 22455 + }, + { + "epoch": 1.0974275034812988, + "grad_norm": 0.0012385837035253644, + "learning_rate": 0.12121034137841868, + "loss": 0.0827, + "num_input_tokens_seen": 29123808, + "step": 22460 + }, + { + "epoch": 1.097671805144993, + "grad_norm": 0.0008596571860834956, + "learning_rate": 0.12115253387250258, + "loss": 0.0831, + "num_input_tokens_seen": 29130528, + "step": 22465 + }, + { + "epoch": 1.0979161068086873, + "grad_norm": 0.0012843571603298187, + "learning_rate": 0.12109473081522831, + "loss": 0.1105, + "num_input_tokens_seen": 29137408, + "step": 22470 + }, + { + "epoch": 1.0981604084723817, + "grad_norm": 0.002025584690272808, + "learning_rate": 0.12103693221550982, + "loss": 0.0666, + "num_input_tokens_seen": 29143680, + "step": 22475 + }, + { + "epoch": 1.098404710136076, + "grad_norm": 0.0014435274060815573, + "learning_rate": 0.12097913808226027, + "loss": 0.1021, + "num_input_tokens_seen": 29150432, + "step": 22480 + }, + { + "epoch": 1.0986490117997703, + "grad_norm": 0.0018563710618764162, + "learning_rate": 0.12092134842439234, + "loss": 0.0867, + "num_input_tokens_seen": 29156768, + "step": 22485 + }, + { + "epoch": 1.0988933134634646, + "grad_norm": 0.001007911516353488, + "learning_rate": 0.12086356325081798, + "loss": 0.0892, + "num_input_tokens_seen": 29163168, + "step": 22490 + }, + { + "epoch": 1.099137615127159, + "grad_norm": 0.000931190385017544, + "learning_rate": 0.12080578257044824, + "loss": 0.0697, + "num_input_tokens_seen": 29170432, + "step": 22495 + }, + { + "epoch": 1.0993819167908534, + "grad_norm": 0.0011492520570755005, + "learning_rate": 0.12074800639219378, + "loss": 0.089, + "num_input_tokens_seen": 29177568, + "step": 22500 + }, + { + "epoch": 1.0996262184545478, + "grad_norm": 0.0008479716489091516, + "learning_rate": 0.12069023472496428, + "loss": 0.0595, + "num_input_tokens_seen": 29184288, + "step": 22505 + }, + { + "epoch": 1.099870520118242, + "grad_norm": 0.001032893662340939, + "learning_rate": 0.12063246757766893, + "loss": 0.0684, + "num_input_tokens_seen": 29190720, + "step": 22510 + }, + { + "epoch": 1.1001148217819363, + "grad_norm": 0.0010732300579547882, + "learning_rate": 0.12057470495921618, + "loss": 0.1052, + "num_input_tokens_seen": 29196768, + "step": 22515 + }, + { + "epoch": 1.1003591234456307, + "grad_norm": 0.001476700883358717, + "learning_rate": 0.12051694687851364, + "loss": 0.0787, + "num_input_tokens_seen": 29203072, + "step": 22520 + }, + { + "epoch": 1.100603425109325, + "grad_norm": 0.0010828949743881822, + "learning_rate": 0.12045919334446839, + "loss": 0.0761, + "num_input_tokens_seen": 29209344, + "step": 22525 + }, + { + "epoch": 1.1008477267730192, + "grad_norm": 0.0009777280502021313, + "learning_rate": 0.12040144436598683, + "loss": 0.1014, + "num_input_tokens_seen": 29216064, + "step": 22530 + }, + { + "epoch": 1.1010920284367136, + "grad_norm": 0.0009147629607468843, + "learning_rate": 0.12034369995197444, + "loss": 0.0765, + "num_input_tokens_seen": 29222528, + "step": 22535 + }, + { + "epoch": 1.101336330100408, + "grad_norm": 0.0026257222052663565, + "learning_rate": 0.12028596011133627, + "loss": 0.1049, + "num_input_tokens_seen": 29228864, + "step": 22540 + }, + { + "epoch": 1.1015806317641024, + "grad_norm": 0.0008910034084692597, + "learning_rate": 0.12022822485297643, + "loss": 0.077, + "num_input_tokens_seen": 29235168, + "step": 22545 + }, + { + "epoch": 1.1018249334277965, + "grad_norm": 0.001612793537788093, + "learning_rate": 0.12017049418579843, + "loss": 0.108, + "num_input_tokens_seen": 29241408, + "step": 22550 + }, + { + "epoch": 1.102069235091491, + "grad_norm": 0.0012279574293643236, + "learning_rate": 0.12011276811870514, + "loss": 0.0869, + "num_input_tokens_seen": 29248384, + "step": 22555 + }, + { + "epoch": 1.1023135367551853, + "grad_norm": 0.0016819167649373412, + "learning_rate": 0.12005504666059852, + "loss": 0.0872, + "num_input_tokens_seen": 29255104, + "step": 22560 + }, + { + "epoch": 1.1025578384188797, + "grad_norm": 0.00108810281381011, + "learning_rate": 0.11999732982038003, + "loss": 0.099, + "num_input_tokens_seen": 29261568, + "step": 22565 + }, + { + "epoch": 1.102802140082574, + "grad_norm": 0.0005232211551629007, + "learning_rate": 0.11993961760695038, + "loss": 0.0713, + "num_input_tokens_seen": 29268000, + "step": 22570 + }, + { + "epoch": 1.1030464417462682, + "grad_norm": 0.0009668603306636214, + "learning_rate": 0.11988191002920942, + "loss": 0.0938, + "num_input_tokens_seen": 29274016, + "step": 22575 + }, + { + "epoch": 1.1032907434099626, + "grad_norm": 0.0014998686965554953, + "learning_rate": 0.11982420709605641, + "loss": 0.067, + "num_input_tokens_seen": 29280704, + "step": 22580 + }, + { + "epoch": 1.103535045073657, + "grad_norm": 0.0012822216376662254, + "learning_rate": 0.11976650881638991, + "loss": 0.0839, + "num_input_tokens_seen": 29287168, + "step": 22585 + }, + { + "epoch": 1.1037793467373513, + "grad_norm": 0.0009978878078982234, + "learning_rate": 0.11970881519910764, + "loss": 0.0739, + "num_input_tokens_seen": 29293280, + "step": 22590 + }, + { + "epoch": 1.1040236484010455, + "grad_norm": 0.0008640151936560869, + "learning_rate": 0.1196511262531068, + "loss": 0.1089, + "num_input_tokens_seen": 29299520, + "step": 22595 + }, + { + "epoch": 1.1042679500647399, + "grad_norm": 0.0009699521469883621, + "learning_rate": 0.11959344198728361, + "loss": 0.1056, + "num_input_tokens_seen": 29306368, + "step": 22600 + }, + { + "epoch": 1.1042679500647399, + "eval_loss": 0.0956297293305397, + "eval_runtime": 401.8053, + "eval_samples_per_second": 90.554, + "eval_steps_per_second": 22.64, + "num_input_tokens_seen": 29306368, + "step": 22600 + }, + { + "epoch": 1.1045122517284343, + "grad_norm": 0.001815083553083241, + "learning_rate": 0.11953576241053378, + "loss": 0.0575, + "num_input_tokens_seen": 29313408, + "step": 22605 + }, + { + "epoch": 1.1047565533921286, + "grad_norm": 0.0016144990222528577, + "learning_rate": 0.11947808753175228, + "loss": 0.0934, + "num_input_tokens_seen": 29320224, + "step": 22610 + }, + { + "epoch": 1.105000855055823, + "grad_norm": 0.002225106582045555, + "learning_rate": 0.1194204173598332, + "loss": 0.0806, + "num_input_tokens_seen": 29326848, + "step": 22615 + }, + { + "epoch": 1.1052451567195172, + "grad_norm": 0.001071340637281537, + "learning_rate": 0.11936275190367007, + "loss": 0.1031, + "num_input_tokens_seen": 29333632, + "step": 22620 + }, + { + "epoch": 1.1054894583832116, + "grad_norm": 0.002150610787793994, + "learning_rate": 0.11930509117215563, + "loss": 0.1226, + "num_input_tokens_seen": 29340000, + "step": 22625 + }, + { + "epoch": 1.105733760046906, + "grad_norm": 0.001468008034862578, + "learning_rate": 0.11924743517418179, + "loss": 0.0969, + "num_input_tokens_seen": 29346368, + "step": 22630 + }, + { + "epoch": 1.1059780617106003, + "grad_norm": 0.000911967596039176, + "learning_rate": 0.11918978391864, + "loss": 0.0919, + "num_input_tokens_seen": 29352736, + "step": 22635 + }, + { + "epoch": 1.1062223633742945, + "grad_norm": 0.0014561722055077553, + "learning_rate": 0.11913213741442065, + "loss": 0.0939, + "num_input_tokens_seen": 29358912, + "step": 22640 + }, + { + "epoch": 1.1064666650379889, + "grad_norm": 0.001115082181058824, + "learning_rate": 0.11907449567041364, + "loss": 0.0961, + "num_input_tokens_seen": 29365248, + "step": 22645 + }, + { + "epoch": 1.1067109667016832, + "grad_norm": 0.0009365850710310042, + "learning_rate": 0.11901685869550803, + "loss": 0.0764, + "num_input_tokens_seen": 29371936, + "step": 22650 + }, + { + "epoch": 1.1069552683653776, + "grad_norm": 0.0013738427078351378, + "learning_rate": 0.1189592264985922, + "loss": 0.0729, + "num_input_tokens_seen": 29378432, + "step": 22655 + }, + { + "epoch": 1.107199570029072, + "grad_norm": 0.0008525936282239854, + "learning_rate": 0.11890159908855373, + "loss": 0.0897, + "num_input_tokens_seen": 29384864, + "step": 22660 + }, + { + "epoch": 1.1074438716927661, + "grad_norm": 0.0016903226496651769, + "learning_rate": 0.11884397647427941, + "loss": 0.1049, + "num_input_tokens_seen": 29391264, + "step": 22665 + }, + { + "epoch": 1.1076881733564605, + "grad_norm": 0.0013549641007557511, + "learning_rate": 0.11878635866465546, + "loss": 0.0821, + "num_input_tokens_seen": 29397568, + "step": 22670 + }, + { + "epoch": 1.107932475020155, + "grad_norm": 0.0018646264215931296, + "learning_rate": 0.11872874566856734, + "loss": 0.0786, + "num_input_tokens_seen": 29404672, + "step": 22675 + }, + { + "epoch": 1.1081767766838493, + "grad_norm": 0.0016617337241768837, + "learning_rate": 0.11867113749489955, + "loss": 0.08, + "num_input_tokens_seen": 29410816, + "step": 22680 + }, + { + "epoch": 1.1084210783475434, + "grad_norm": 0.0009732077596709132, + "learning_rate": 0.11861353415253607, + "loss": 0.1216, + "num_input_tokens_seen": 29417312, + "step": 22685 + }, + { + "epoch": 1.1086653800112378, + "grad_norm": 0.0008729675319045782, + "learning_rate": 0.11855593565036011, + "loss": 0.0786, + "num_input_tokens_seen": 29424000, + "step": 22690 + }, + { + "epoch": 1.1089096816749322, + "grad_norm": 0.0008465920109301805, + "learning_rate": 0.11849834199725394, + "loss": 0.0804, + "num_input_tokens_seen": 29430080, + "step": 22695 + }, + { + "epoch": 1.1091539833386266, + "grad_norm": 0.0012903816532343626, + "learning_rate": 0.1184407532020994, + "loss": 0.0817, + "num_input_tokens_seen": 29436096, + "step": 22700 + }, + { + "epoch": 1.109398285002321, + "grad_norm": 0.0019492337014526129, + "learning_rate": 0.11838316927377723, + "loss": 0.07, + "num_input_tokens_seen": 29442688, + "step": 22705 + }, + { + "epoch": 1.1096425866660151, + "grad_norm": 0.00226630293764174, + "learning_rate": 0.11832559022116766, + "loss": 0.1123, + "num_input_tokens_seen": 29449280, + "step": 22710 + }, + { + "epoch": 1.1098868883297095, + "grad_norm": 0.0009272925090044737, + "learning_rate": 0.11826801605315022, + "loss": 0.0675, + "num_input_tokens_seen": 29455488, + "step": 22715 + }, + { + "epoch": 1.1101311899934039, + "grad_norm": 0.001682823640294373, + "learning_rate": 0.1182104467786034, + "loss": 0.0897, + "num_input_tokens_seen": 29462400, + "step": 22720 + }, + { + "epoch": 1.1103754916570983, + "grad_norm": 0.0009082207689061761, + "learning_rate": 0.1181528824064052, + "loss": 0.1009, + "num_input_tokens_seen": 29468832, + "step": 22725 + }, + { + "epoch": 1.1106197933207924, + "grad_norm": 0.0011318254983052611, + "learning_rate": 0.11809532294543279, + "loss": 0.0984, + "num_input_tokens_seen": 29475456, + "step": 22730 + }, + { + "epoch": 1.1108640949844868, + "grad_norm": 0.0017432866152375937, + "learning_rate": 0.11803776840456245, + "loss": 0.0955, + "num_input_tokens_seen": 29482336, + "step": 22735 + }, + { + "epoch": 1.1111083966481812, + "grad_norm": 0.0013830552343279123, + "learning_rate": 0.11798021879266997, + "loss": 0.1116, + "num_input_tokens_seen": 29488800, + "step": 22740 + }, + { + "epoch": 1.1113526983118756, + "grad_norm": 0.0011635273694992065, + "learning_rate": 0.11792267411863006, + "loss": 0.1018, + "num_input_tokens_seen": 29495264, + "step": 22745 + }, + { + "epoch": 1.11159699997557, + "grad_norm": 0.0012137737357988954, + "learning_rate": 0.1178651343913169, + "loss": 0.0697, + "num_input_tokens_seen": 29501632, + "step": 22750 + }, + { + "epoch": 1.111841301639264, + "grad_norm": 0.0009532968979328871, + "learning_rate": 0.11780759961960392, + "loss": 0.085, + "num_input_tokens_seen": 29508096, + "step": 22755 + }, + { + "epoch": 1.1120856033029585, + "grad_norm": 0.0016980640357360244, + "learning_rate": 0.1177500698123636, + "loss": 0.1114, + "num_input_tokens_seen": 29514976, + "step": 22760 + }, + { + "epoch": 1.1123299049666528, + "grad_norm": 0.0011556892422959208, + "learning_rate": 0.11769254497846778, + "loss": 0.0854, + "num_input_tokens_seen": 29521312, + "step": 22765 + }, + { + "epoch": 1.1125742066303472, + "grad_norm": 0.0009634310845285654, + "learning_rate": 0.11763502512678758, + "loss": 0.0771, + "num_input_tokens_seen": 29527872, + "step": 22770 + }, + { + "epoch": 1.1128185082940414, + "grad_norm": 0.0012828842736780643, + "learning_rate": 0.11757751026619315, + "loss": 0.0799, + "num_input_tokens_seen": 29534944, + "step": 22775 + }, + { + "epoch": 1.1130628099577358, + "grad_norm": 0.0010299670975655317, + "learning_rate": 0.11752000040555416, + "loss": 0.1157, + "num_input_tokens_seen": 29541696, + "step": 22780 + }, + { + "epoch": 1.1133071116214301, + "grad_norm": 0.0006981150363571942, + "learning_rate": 0.11746249555373921, + "loss": 0.069, + "num_input_tokens_seen": 29548128, + "step": 22785 + }, + { + "epoch": 1.1135514132851245, + "grad_norm": 0.0008846033015288413, + "learning_rate": 0.11740499571961638, + "loss": 0.0919, + "num_input_tokens_seen": 29554656, + "step": 22790 + }, + { + "epoch": 1.113795714948819, + "grad_norm": 0.0009987849043682218, + "learning_rate": 0.11734750091205279, + "loss": 0.1102, + "num_input_tokens_seen": 29561024, + "step": 22795 + }, + { + "epoch": 1.114040016612513, + "grad_norm": 0.0006589735276065767, + "learning_rate": 0.11729001113991493, + "loss": 0.0945, + "num_input_tokens_seen": 29567616, + "step": 22800 + }, + { + "epoch": 1.114040016612513, + "eval_loss": 0.09384085237979889, + "eval_runtime": 402.3772, + "eval_samples_per_second": 90.425, + "eval_steps_per_second": 22.608, + "num_input_tokens_seen": 29567616, + "step": 22800 + }, + { + "epoch": 1.1142843182762074, + "grad_norm": 0.0006671895971521735, + "learning_rate": 0.11723252641206837, + "loss": 0.0839, + "num_input_tokens_seen": 29573824, + "step": 22805 + }, + { + "epoch": 1.1145286199399018, + "grad_norm": 0.0009115938446484506, + "learning_rate": 0.11717504673737808, + "loss": 0.0659, + "num_input_tokens_seen": 29580096, + "step": 22810 + }, + { + "epoch": 1.1147729216035962, + "grad_norm": 0.001443873276002705, + "learning_rate": 0.11711757212470802, + "loss": 0.1305, + "num_input_tokens_seen": 29586880, + "step": 22815 + }, + { + "epoch": 1.1150172232672904, + "grad_norm": 0.0008223216864280403, + "learning_rate": 0.11706010258292165, + "loss": 0.0839, + "num_input_tokens_seen": 29593728, + "step": 22820 + }, + { + "epoch": 1.1152615249309847, + "grad_norm": 0.001257190597243607, + "learning_rate": 0.11700263812088131, + "loss": 0.0791, + "num_input_tokens_seen": 29600544, + "step": 22825 + }, + { + "epoch": 1.1155058265946791, + "grad_norm": 0.0018495457479730248, + "learning_rate": 0.11694517874744892, + "loss": 0.0793, + "num_input_tokens_seen": 29607136, + "step": 22830 + }, + { + "epoch": 1.1157501282583735, + "grad_norm": 0.0011677911970764399, + "learning_rate": 0.11688772447148532, + "loss": 0.1245, + "num_input_tokens_seen": 29613184, + "step": 22835 + }, + { + "epoch": 1.1159944299220679, + "grad_norm": 0.0013172619510442019, + "learning_rate": 0.11683027530185074, + "loss": 0.0933, + "num_input_tokens_seen": 29619424, + "step": 22840 + }, + { + "epoch": 1.116238731585762, + "grad_norm": 0.0010473121656104922, + "learning_rate": 0.11677283124740451, + "loss": 0.0944, + "num_input_tokens_seen": 29626208, + "step": 22845 + }, + { + "epoch": 1.1164830332494564, + "grad_norm": 0.0007953434251248837, + "learning_rate": 0.11671539231700531, + "loss": 0.0836, + "num_input_tokens_seen": 29632640, + "step": 22850 + }, + { + "epoch": 1.1167273349131508, + "grad_norm": 0.0009367679012939334, + "learning_rate": 0.11665795851951084, + "loss": 0.07, + "num_input_tokens_seen": 29640032, + "step": 22855 + }, + { + "epoch": 1.1169716365768452, + "grad_norm": 0.001858372357673943, + "learning_rate": 0.11660052986377825, + "loss": 0.0788, + "num_input_tokens_seen": 29646272, + "step": 22860 + }, + { + "epoch": 1.1172159382405393, + "grad_norm": 0.0015237312763929367, + "learning_rate": 0.1165431063586636, + "loss": 0.0853, + "num_input_tokens_seen": 29652576, + "step": 22865 + }, + { + "epoch": 1.1174602399042337, + "grad_norm": 0.0011607040651142597, + "learning_rate": 0.11648568801302245, + "loss": 0.0803, + "num_input_tokens_seen": 29659616, + "step": 22870 + }, + { + "epoch": 1.117704541567928, + "grad_norm": 0.002002204302698374, + "learning_rate": 0.11642827483570937, + "loss": 0.0788, + "num_input_tokens_seen": 29666272, + "step": 22875 + }, + { + "epoch": 1.1179488432316225, + "grad_norm": 0.0010703313164412975, + "learning_rate": 0.11637086683557815, + "loss": 0.0811, + "num_input_tokens_seen": 29672544, + "step": 22880 + }, + { + "epoch": 1.1181931448953168, + "grad_norm": 0.0009873281233012676, + "learning_rate": 0.11631346402148188, + "loss": 0.0687, + "num_input_tokens_seen": 29679168, + "step": 22885 + }, + { + "epoch": 1.118437446559011, + "grad_norm": 0.0012869761558249593, + "learning_rate": 0.11625606640227285, + "loss": 0.0705, + "num_input_tokens_seen": 29686176, + "step": 22890 + }, + { + "epoch": 1.1186817482227054, + "grad_norm": 0.0013319561257958412, + "learning_rate": 0.11619867398680238, + "loss": 0.064, + "num_input_tokens_seen": 29693120, + "step": 22895 + }, + { + "epoch": 1.1189260498863998, + "grad_norm": 0.0011749890400096774, + "learning_rate": 0.11614128678392119, + "loss": 0.0804, + "num_input_tokens_seen": 29698912, + "step": 22900 + }, + { + "epoch": 1.1191703515500941, + "grad_norm": 0.0008904912392608821, + "learning_rate": 0.11608390480247906, + "loss": 0.094, + "num_input_tokens_seen": 29705056, + "step": 22905 + }, + { + "epoch": 1.1194146532137883, + "grad_norm": 0.001070785103365779, + "learning_rate": 0.11602652805132499, + "loss": 0.1022, + "num_input_tokens_seen": 29711936, + "step": 22910 + }, + { + "epoch": 1.1196589548774827, + "grad_norm": 0.0018644554074853659, + "learning_rate": 0.11596915653930731, + "loss": 0.1002, + "num_input_tokens_seen": 29718816, + "step": 22915 + }, + { + "epoch": 1.119903256541177, + "grad_norm": 0.0007122614188119769, + "learning_rate": 0.11591179027527328, + "loss": 0.0853, + "num_input_tokens_seen": 29725472, + "step": 22920 + }, + { + "epoch": 1.1201475582048714, + "grad_norm": 0.0013772232923656702, + "learning_rate": 0.11585442926806956, + "loss": 0.0975, + "num_input_tokens_seen": 29732384, + "step": 22925 + }, + { + "epoch": 1.1203918598685658, + "grad_norm": 0.0018739770166575909, + "learning_rate": 0.11579707352654202, + "loss": 0.0936, + "num_input_tokens_seen": 29738880, + "step": 22930 + }, + { + "epoch": 1.12063616153226, + "grad_norm": 0.0010754610411822796, + "learning_rate": 0.11573972305953548, + "loss": 0.1061, + "num_input_tokens_seen": 29745088, + "step": 22935 + }, + { + "epoch": 1.1208804631959544, + "grad_norm": 0.0011936796363443136, + "learning_rate": 0.11568237787589426, + "loss": 0.0793, + "num_input_tokens_seen": 29751936, + "step": 22940 + }, + { + "epoch": 1.1211247648596487, + "grad_norm": 0.0011385231046006083, + "learning_rate": 0.11562503798446161, + "loss": 0.1013, + "num_input_tokens_seen": 29758304, + "step": 22945 + }, + { + "epoch": 1.1213690665233431, + "grad_norm": 0.0013281998690217733, + "learning_rate": 0.11556770339408005, + "loss": 0.0988, + "num_input_tokens_seen": 29764800, + "step": 22950 + }, + { + "epoch": 1.1216133681870373, + "grad_norm": 0.0009100367315113544, + "learning_rate": 0.1155103741135914, + "loss": 0.101, + "num_input_tokens_seen": 29770848, + "step": 22955 + }, + { + "epoch": 1.1218576698507317, + "grad_norm": 0.0013085377868264914, + "learning_rate": 0.1154530501518364, + "loss": 0.1027, + "num_input_tokens_seen": 29777568, + "step": 22960 + }, + { + "epoch": 1.122101971514426, + "grad_norm": 0.0010880501940846443, + "learning_rate": 0.11539573151765523, + "loss": 0.0929, + "num_input_tokens_seen": 29784160, + "step": 22965 + }, + { + "epoch": 1.1223462731781204, + "grad_norm": 0.0013809552183374763, + "learning_rate": 0.11533841821988719, + "loss": 0.1007, + "num_input_tokens_seen": 29790400, + "step": 22970 + }, + { + "epoch": 1.1225905748418146, + "grad_norm": 0.0005781057989224792, + "learning_rate": 0.11528111026737059, + "loss": 0.1005, + "num_input_tokens_seen": 29796672, + "step": 22975 + }, + { + "epoch": 1.122834876505509, + "grad_norm": 0.0014771645655855536, + "learning_rate": 0.11522380766894312, + "loss": 0.1097, + "num_input_tokens_seen": 29803232, + "step": 22980 + }, + { + "epoch": 1.1230791781692033, + "grad_norm": 0.0008638349245302379, + "learning_rate": 0.11516651043344152, + "loss": 0.0916, + "num_input_tokens_seen": 29810048, + "step": 22985 + }, + { + "epoch": 1.1233234798328977, + "grad_norm": 0.00121233647223562, + "learning_rate": 0.11510921856970172, + "loss": 0.1031, + "num_input_tokens_seen": 29816640, + "step": 22990 + }, + { + "epoch": 1.123567781496592, + "grad_norm": 0.0012830840423703194, + "learning_rate": 0.11505193208655895, + "loss": 0.1064, + "num_input_tokens_seen": 29823200, + "step": 22995 + }, + { + "epoch": 1.1238120831602862, + "grad_norm": 0.000694166636094451, + "learning_rate": 0.11499465099284738, + "loss": 0.0968, + "num_input_tokens_seen": 29829920, + "step": 23000 + }, + { + "epoch": 1.1238120831602862, + "eval_loss": 0.09254075586795807, + "eval_runtime": 402.4675, + "eval_samples_per_second": 90.405, + "eval_steps_per_second": 22.603, + "num_input_tokens_seen": 29829920, + "step": 23000 + }, + { + "epoch": 1.1240563848239806, + "grad_norm": 0.0008355005411431193, + "learning_rate": 0.1149373752974006, + "loss": 0.07, + "num_input_tokens_seen": 29836448, + "step": 23005 + }, + { + "epoch": 1.124300686487675, + "grad_norm": 0.0008243769407272339, + "learning_rate": 0.11488010500905109, + "loss": 0.1059, + "num_input_tokens_seen": 29842816, + "step": 23010 + }, + { + "epoch": 1.1245449881513694, + "grad_norm": 0.0012553032720461488, + "learning_rate": 0.11482284013663077, + "loss": 0.0858, + "num_input_tokens_seen": 29849344, + "step": 23015 + }, + { + "epoch": 1.1247892898150635, + "grad_norm": 0.0012229477288201451, + "learning_rate": 0.11476558068897061, + "loss": 0.0916, + "num_input_tokens_seen": 29855680, + "step": 23020 + }, + { + "epoch": 1.125033591478758, + "grad_norm": 0.0011784235248342156, + "learning_rate": 0.11470832667490061, + "loss": 0.0848, + "num_input_tokens_seen": 29862304, + "step": 23025 + }, + { + "epoch": 1.1252778931424523, + "grad_norm": 0.0012178502511233091, + "learning_rate": 0.11465107810325013, + "loss": 0.0787, + "num_input_tokens_seen": 29869312, + "step": 23030 + }, + { + "epoch": 1.1255221948061467, + "grad_norm": 0.0013691347558051348, + "learning_rate": 0.11459383498284771, + "loss": 0.1007, + "num_input_tokens_seen": 29875712, + "step": 23035 + }, + { + "epoch": 1.1257664964698408, + "grad_norm": 0.0017746285302564502, + "learning_rate": 0.11453659732252082, + "loss": 0.1041, + "num_input_tokens_seen": 29882176, + "step": 23040 + }, + { + "epoch": 1.1260107981335352, + "grad_norm": 0.0010581579990684986, + "learning_rate": 0.11447936513109633, + "loss": 0.0762, + "num_input_tokens_seen": 29888800, + "step": 23045 + }, + { + "epoch": 1.1262550997972296, + "grad_norm": 0.0011982058640569448, + "learning_rate": 0.11442213841740011, + "loss": 0.0709, + "num_input_tokens_seen": 29895264, + "step": 23050 + }, + { + "epoch": 1.126499401460924, + "grad_norm": 0.0006203183438628912, + "learning_rate": 0.1143649171902572, + "loss": 0.0697, + "num_input_tokens_seen": 29901760, + "step": 23055 + }, + { + "epoch": 1.1267437031246184, + "grad_norm": 0.0014560542767867446, + "learning_rate": 0.11430770145849194, + "loss": 0.0913, + "num_input_tokens_seen": 29908896, + "step": 23060 + }, + { + "epoch": 1.1269880047883125, + "grad_norm": 0.0016031808918341994, + "learning_rate": 0.11425049123092756, + "loss": 0.1033, + "num_input_tokens_seen": 29915296, + "step": 23065 + }, + { + "epoch": 1.127232306452007, + "grad_norm": 0.001249050605110824, + "learning_rate": 0.11419328651638674, + "loss": 0.0892, + "num_input_tokens_seen": 29921280, + "step": 23070 + }, + { + "epoch": 1.1274766081157013, + "grad_norm": 0.0010600927053019404, + "learning_rate": 0.11413608732369115, + "loss": 0.0798, + "num_input_tokens_seen": 29928416, + "step": 23075 + }, + { + "epoch": 1.1277209097793957, + "grad_norm": 0.0014771948335692286, + "learning_rate": 0.11407889366166153, + "loss": 0.1022, + "num_input_tokens_seen": 29934976, + "step": 23080 + }, + { + "epoch": 1.1279652114430898, + "grad_norm": 0.0014995633391663432, + "learning_rate": 0.11402170553911797, + "loss": 0.0879, + "num_input_tokens_seen": 29941472, + "step": 23085 + }, + { + "epoch": 1.1282095131067842, + "grad_norm": 0.0011194770922884345, + "learning_rate": 0.11396452296487955, + "loss": 0.0675, + "num_input_tokens_seen": 29948352, + "step": 23090 + }, + { + "epoch": 1.1284538147704786, + "grad_norm": 0.0008845528936944902, + "learning_rate": 0.11390734594776449, + "loss": 0.0857, + "num_input_tokens_seen": 29954976, + "step": 23095 + }, + { + "epoch": 1.128698116434173, + "grad_norm": 0.0014972048811614513, + "learning_rate": 0.11385017449659031, + "loss": 0.1047, + "num_input_tokens_seen": 29961504, + "step": 23100 + }, + { + "epoch": 1.1289424180978673, + "grad_norm": 0.0010997607605531812, + "learning_rate": 0.11379300862017344, + "loss": 0.0804, + "num_input_tokens_seen": 29967936, + "step": 23105 + }, + { + "epoch": 1.1291867197615615, + "grad_norm": 0.0014192858943715692, + "learning_rate": 0.11373584832732966, + "loss": 0.0962, + "num_input_tokens_seen": 29974240, + "step": 23110 + }, + { + "epoch": 1.1294310214252559, + "grad_norm": 0.0019656866788864136, + "learning_rate": 0.11367869362687386, + "loss": 0.1086, + "num_input_tokens_seen": 29980608, + "step": 23115 + }, + { + "epoch": 1.1296753230889502, + "grad_norm": 0.001459824969060719, + "learning_rate": 0.11362154452761988, + "loss": 0.0893, + "num_input_tokens_seen": 29986624, + "step": 23120 + }, + { + "epoch": 1.1299196247526446, + "grad_norm": 0.0015809257747605443, + "learning_rate": 0.11356440103838095, + "loss": 0.1031, + "num_input_tokens_seen": 29993280, + "step": 23125 + }, + { + "epoch": 1.1301639264163388, + "grad_norm": 0.0013096017064526677, + "learning_rate": 0.11350726316796922, + "loss": 0.1169, + "num_input_tokens_seen": 29999840, + "step": 23130 + }, + { + "epoch": 1.1304082280800332, + "grad_norm": 0.0012436212273314595, + "learning_rate": 0.11345013092519607, + "loss": 0.0879, + "num_input_tokens_seen": 30006304, + "step": 23135 + }, + { + "epoch": 1.1306525297437275, + "grad_norm": 0.0015291267773136497, + "learning_rate": 0.11339300431887213, + "loss": 0.089, + "num_input_tokens_seen": 30012928, + "step": 23140 + }, + { + "epoch": 1.130896831407422, + "grad_norm": 0.0019381640013307333, + "learning_rate": 0.11333588335780687, + "loss": 0.0669, + "num_input_tokens_seen": 30019264, + "step": 23145 + }, + { + "epoch": 1.1311411330711163, + "grad_norm": 0.001631527324207127, + "learning_rate": 0.11327876805080916, + "loss": 0.1014, + "num_input_tokens_seen": 30025504, + "step": 23150 + }, + { + "epoch": 1.1313854347348105, + "grad_norm": 0.0008962780702859163, + "learning_rate": 0.11322165840668696, + "loss": 0.0799, + "num_input_tokens_seen": 30032096, + "step": 23155 + }, + { + "epoch": 1.1316297363985048, + "grad_norm": 0.0009562247432768345, + "learning_rate": 0.11316455443424717, + "loss": 0.0772, + "num_input_tokens_seen": 30038400, + "step": 23160 + }, + { + "epoch": 1.1318740380621992, + "grad_norm": 0.0023482993710786104, + "learning_rate": 0.11310745614229603, + "loss": 0.1033, + "num_input_tokens_seen": 30045024, + "step": 23165 + }, + { + "epoch": 1.1321183397258936, + "grad_norm": 0.0009518415317870677, + "learning_rate": 0.1130503635396387, + "loss": 0.0928, + "num_input_tokens_seen": 30051552, + "step": 23170 + }, + { + "epoch": 1.1323626413895878, + "grad_norm": 0.001949106459505856, + "learning_rate": 0.11299327663507966, + "loss": 0.0807, + "num_input_tokens_seen": 30058688, + "step": 23175 + }, + { + "epoch": 1.1326069430532821, + "grad_norm": 0.001288347295485437, + "learning_rate": 0.11293619543742246, + "loss": 0.0706, + "num_input_tokens_seen": 30065760, + "step": 23180 + }, + { + "epoch": 1.1328512447169765, + "grad_norm": 0.0011806429829448462, + "learning_rate": 0.11287911995546965, + "loss": 0.0811, + "num_input_tokens_seen": 30072160, + "step": 23185 + }, + { + "epoch": 1.133095546380671, + "grad_norm": 0.0009797160746529698, + "learning_rate": 0.11282205019802308, + "loss": 0.084, + "num_input_tokens_seen": 30078720, + "step": 23190 + }, + { + "epoch": 1.1333398480443653, + "grad_norm": 0.0010089566931128502, + "learning_rate": 0.11276498617388354, + "loss": 0.0993, + "num_input_tokens_seen": 30085888, + "step": 23195 + }, + { + "epoch": 1.1335841497080594, + "grad_norm": 0.0013945710379630327, + "learning_rate": 0.11270792789185109, + "loss": 0.0798, + "num_input_tokens_seen": 30092128, + "step": 23200 + }, + { + "epoch": 1.1335841497080594, + "eval_loss": 0.09246733039617538, + "eval_runtime": 401.9381, + "eval_samples_per_second": 90.524, + "eval_steps_per_second": 22.633, + "num_input_tokens_seen": 30092128, + "step": 23200 + }, + { + "epoch": 1.1338284513717538, + "grad_norm": 0.0010442690690979362, + "learning_rate": 0.11265087536072482, + "loss": 0.0822, + "num_input_tokens_seen": 30098848, + "step": 23205 + }, + { + "epoch": 1.1340727530354482, + "grad_norm": 0.001672646147198975, + "learning_rate": 0.11259382858930288, + "loss": 0.1009, + "num_input_tokens_seen": 30105344, + "step": 23210 + }, + { + "epoch": 1.1343170546991426, + "grad_norm": 0.0012657458428293467, + "learning_rate": 0.11253678758638262, + "loss": 0.0909, + "num_input_tokens_seen": 30111680, + "step": 23215 + }, + { + "epoch": 1.1345613563628367, + "grad_norm": 0.001002528122626245, + "learning_rate": 0.11247975236076059, + "loss": 0.0927, + "num_input_tokens_seen": 30118464, + "step": 23220 + }, + { + "epoch": 1.134805658026531, + "grad_norm": 0.0010346167255192995, + "learning_rate": 0.11242272292123218, + "loss": 0.084, + "num_input_tokens_seen": 30124704, + "step": 23225 + }, + { + "epoch": 1.1350499596902255, + "grad_norm": 0.0010710906935855746, + "learning_rate": 0.11236569927659217, + "loss": 0.0804, + "num_input_tokens_seen": 30130816, + "step": 23230 + }, + { + "epoch": 1.1352942613539199, + "grad_norm": 0.003068498568609357, + "learning_rate": 0.11230868143563429, + "loss": 0.0967, + "num_input_tokens_seen": 30137408, + "step": 23235 + }, + { + "epoch": 1.1355385630176142, + "grad_norm": 0.0014878022484481335, + "learning_rate": 0.11225166940715131, + "loss": 0.0852, + "num_input_tokens_seen": 30144160, + "step": 23240 + }, + { + "epoch": 1.1357828646813084, + "grad_norm": 0.0008877896470949054, + "learning_rate": 0.11219466319993537, + "loss": 0.1135, + "num_input_tokens_seen": 30150464, + "step": 23245 + }, + { + "epoch": 1.1360271663450028, + "grad_norm": 0.0014459833037108183, + "learning_rate": 0.11213766282277739, + "loss": 0.0924, + "num_input_tokens_seen": 30156768, + "step": 23250 + }, + { + "epoch": 1.1362714680086972, + "grad_norm": 0.0009936465648934245, + "learning_rate": 0.11208066828446761, + "loss": 0.0748, + "num_input_tokens_seen": 30163776, + "step": 23255 + }, + { + "epoch": 1.1365157696723915, + "grad_norm": 0.0011059241369366646, + "learning_rate": 0.11202367959379537, + "loss": 0.0981, + "num_input_tokens_seen": 30170240, + "step": 23260 + }, + { + "epoch": 1.1367600713360857, + "grad_norm": 0.0006793513312004507, + "learning_rate": 0.11196669675954894, + "loss": 0.0881, + "num_input_tokens_seen": 30176736, + "step": 23265 + }, + { + "epoch": 1.13700437299978, + "grad_norm": 0.0006532729603350163, + "learning_rate": 0.1119097197905158, + "loss": 0.0875, + "num_input_tokens_seen": 30183104, + "step": 23270 + }, + { + "epoch": 1.1372486746634745, + "grad_norm": 0.0012325322022661567, + "learning_rate": 0.11185274869548259, + "loss": 0.0978, + "num_input_tokens_seen": 30189344, + "step": 23275 + }, + { + "epoch": 1.1374929763271688, + "grad_norm": 0.0013418116141110659, + "learning_rate": 0.11179578348323486, + "loss": 0.0972, + "num_input_tokens_seen": 30195520, + "step": 23280 + }, + { + "epoch": 1.1377372779908632, + "grad_norm": 0.0008867286960594356, + "learning_rate": 0.1117388241625575, + "loss": 0.0829, + "num_input_tokens_seen": 30201632, + "step": 23285 + }, + { + "epoch": 1.1379815796545574, + "grad_norm": 0.0006853225058875978, + "learning_rate": 0.11168187074223421, + "loss": 0.0776, + "num_input_tokens_seen": 30208128, + "step": 23290 + }, + { + "epoch": 1.1382258813182518, + "grad_norm": 0.001610215287655592, + "learning_rate": 0.11162492323104796, + "loss": 0.1114, + "num_input_tokens_seen": 30214144, + "step": 23295 + }, + { + "epoch": 1.1384701829819461, + "grad_norm": 0.0009584352374076843, + "learning_rate": 0.11156798163778091, + "loss": 0.0705, + "num_input_tokens_seen": 30220320, + "step": 23300 + }, + { + "epoch": 1.1387144846456405, + "grad_norm": 0.0038620231207460165, + "learning_rate": 0.11151104597121399, + "loss": 0.085, + "num_input_tokens_seen": 30227200, + "step": 23305 + }, + { + "epoch": 1.1389587863093347, + "grad_norm": 0.00142824521753937, + "learning_rate": 0.11145411624012742, + "loss": 0.082, + "num_input_tokens_seen": 30233632, + "step": 23310 + }, + { + "epoch": 1.139203087973029, + "grad_norm": 0.000946389336604625, + "learning_rate": 0.11139719245330063, + "loss": 0.093, + "num_input_tokens_seen": 30240160, + "step": 23315 + }, + { + "epoch": 1.1394473896367234, + "grad_norm": 0.00174328510183841, + "learning_rate": 0.11134027461951179, + "loss": 0.113, + "num_input_tokens_seen": 30246656, + "step": 23320 + }, + { + "epoch": 1.1396916913004178, + "grad_norm": 0.0017108377069234848, + "learning_rate": 0.11128336274753849, + "loss": 0.0795, + "num_input_tokens_seen": 30253216, + "step": 23325 + }, + { + "epoch": 1.1399359929641122, + "grad_norm": 0.001054993481375277, + "learning_rate": 0.11122645684615715, + "loss": 0.1204, + "num_input_tokens_seen": 30259936, + "step": 23330 + }, + { + "epoch": 1.1401802946278063, + "grad_norm": 0.0007996855420060456, + "learning_rate": 0.11116955692414345, + "loss": 0.0984, + "num_input_tokens_seen": 30266624, + "step": 23335 + }, + { + "epoch": 1.1404245962915007, + "grad_norm": 0.001067113596946001, + "learning_rate": 0.11111266299027203, + "loss": 0.0916, + "num_input_tokens_seen": 30272864, + "step": 23340 + }, + { + "epoch": 1.140668897955195, + "grad_norm": 0.0015758267836645246, + "learning_rate": 0.11105577505331668, + "loss": 0.0884, + "num_input_tokens_seen": 30278976, + "step": 23345 + }, + { + "epoch": 1.1409131996188895, + "grad_norm": 0.001435750164091587, + "learning_rate": 0.11099889312205018, + "loss": 0.0864, + "num_input_tokens_seen": 30285536, + "step": 23350 + }, + { + "epoch": 1.1411575012825836, + "grad_norm": 0.0015367339365184307, + "learning_rate": 0.11094201720524455, + "loss": 0.102, + "num_input_tokens_seen": 30291776, + "step": 23355 + }, + { + "epoch": 1.141401802946278, + "grad_norm": 0.001765075488947332, + "learning_rate": 0.11088514731167064, + "loss": 0.0754, + "num_input_tokens_seen": 30297728, + "step": 23360 + }, + { + "epoch": 1.1416461046099724, + "grad_norm": 0.0010898082982748747, + "learning_rate": 0.11082828345009862, + "loss": 0.0852, + "num_input_tokens_seen": 30304800, + "step": 23365 + }, + { + "epoch": 1.1418904062736668, + "grad_norm": 0.0011013870825991035, + "learning_rate": 0.11077142562929748, + "loss": 0.1088, + "num_input_tokens_seen": 30311136, + "step": 23370 + }, + { + "epoch": 1.1421347079373612, + "grad_norm": 0.0011223952751606703, + "learning_rate": 0.11071457385803554, + "loss": 0.1199, + "num_input_tokens_seen": 30317504, + "step": 23375 + }, + { + "epoch": 1.1423790096010553, + "grad_norm": 0.0007446063682436943, + "learning_rate": 0.11065772814508001, + "loss": 0.0982, + "num_input_tokens_seen": 30324032, + "step": 23380 + }, + { + "epoch": 1.1426233112647497, + "grad_norm": 0.0008210416999645531, + "learning_rate": 0.11060088849919715, + "loss": 0.083, + "num_input_tokens_seen": 30330048, + "step": 23385 + }, + { + "epoch": 1.142867612928444, + "grad_norm": 0.0007453899597749114, + "learning_rate": 0.11054405492915244, + "loss": 0.0622, + "num_input_tokens_seen": 30336832, + "step": 23390 + }, + { + "epoch": 1.1431119145921385, + "grad_norm": 0.0021302595268934965, + "learning_rate": 0.11048722744371031, + "loss": 0.1016, + "num_input_tokens_seen": 30343360, + "step": 23395 + }, + { + "epoch": 1.1433562162558326, + "grad_norm": 0.0014051655307412148, + "learning_rate": 0.1104304060516342, + "loss": 0.0825, + "num_input_tokens_seen": 30349984, + "step": 23400 + }, + { + "epoch": 1.1433562162558326, + "eval_loss": 0.09408387541770935, + "eval_runtime": 402.2369, + "eval_samples_per_second": 90.457, + "eval_steps_per_second": 22.616, + "num_input_tokens_seen": 30349984, + "step": 23400 + }, + { + "epoch": 1.143600517919527, + "grad_norm": 0.0010580437956377864, + "learning_rate": 0.11037359076168682, + "loss": 0.0749, + "num_input_tokens_seen": 30356416, + "step": 23405 + }, + { + "epoch": 1.1438448195832214, + "grad_norm": 0.0011598618002608418, + "learning_rate": 0.11031678158262966, + "loss": 0.0709, + "num_input_tokens_seen": 30362848, + "step": 23410 + }, + { + "epoch": 1.1440891212469158, + "grad_norm": 0.0009011505171656609, + "learning_rate": 0.11025997852322349, + "loss": 0.0966, + "num_input_tokens_seen": 30369536, + "step": 23415 + }, + { + "epoch": 1.1443334229106101, + "grad_norm": 0.0012104118941351771, + "learning_rate": 0.11020318159222807, + "loss": 0.1041, + "num_input_tokens_seen": 30375840, + "step": 23420 + }, + { + "epoch": 1.1445777245743043, + "grad_norm": 0.0010762227466329932, + "learning_rate": 0.1101463907984021, + "loss": 0.0974, + "num_input_tokens_seen": 30382080, + "step": 23425 + }, + { + "epoch": 1.1448220262379987, + "grad_norm": 0.0009284798288717866, + "learning_rate": 0.11008960615050352, + "loss": 0.109, + "num_input_tokens_seen": 30387872, + "step": 23430 + }, + { + "epoch": 1.145066327901693, + "grad_norm": 0.0008903076522983611, + "learning_rate": 0.11003282765728925, + "loss": 0.099, + "num_input_tokens_seen": 30394272, + "step": 23435 + }, + { + "epoch": 1.1453106295653874, + "grad_norm": 0.0007455542800016701, + "learning_rate": 0.10997605532751518, + "loss": 0.0782, + "num_input_tokens_seen": 30400768, + "step": 23440 + }, + { + "epoch": 1.1455549312290816, + "grad_norm": 0.0009765154682099819, + "learning_rate": 0.1099192891699364, + "loss": 0.0748, + "num_input_tokens_seen": 30407392, + "step": 23445 + }, + { + "epoch": 1.145799232892776, + "grad_norm": 0.0014446181012317538, + "learning_rate": 0.10986252919330687, + "loss": 0.1168, + "num_input_tokens_seen": 30413728, + "step": 23450 + }, + { + "epoch": 1.1460435345564703, + "grad_norm": 0.001292589120566845, + "learning_rate": 0.10980577540637973, + "loss": 0.1312, + "num_input_tokens_seen": 30420096, + "step": 23455 + }, + { + "epoch": 1.1462878362201647, + "grad_norm": 0.0008695657597854733, + "learning_rate": 0.10974902781790719, + "loss": 0.0756, + "num_input_tokens_seen": 30426880, + "step": 23460 + }, + { + "epoch": 1.146532137883859, + "grad_norm": 0.0011858546640723944, + "learning_rate": 0.10969228643664032, + "loss": 0.0937, + "num_input_tokens_seen": 30433152, + "step": 23465 + }, + { + "epoch": 1.1467764395475533, + "grad_norm": 0.0021062574815005064, + "learning_rate": 0.10963555127132942, + "loss": 0.0924, + "num_input_tokens_seen": 30439456, + "step": 23470 + }, + { + "epoch": 1.1470207412112476, + "grad_norm": 0.0009893927490338683, + "learning_rate": 0.10957882233072382, + "loss": 0.0919, + "num_input_tokens_seen": 30445536, + "step": 23475 + }, + { + "epoch": 1.147265042874942, + "grad_norm": 0.0006219573551788926, + "learning_rate": 0.10952209962357176, + "loss": 0.0716, + "num_input_tokens_seen": 30451872, + "step": 23480 + }, + { + "epoch": 1.1475093445386364, + "grad_norm": 0.0008460439275950193, + "learning_rate": 0.10946538315862062, + "loss": 0.0887, + "num_input_tokens_seen": 30458400, + "step": 23485 + }, + { + "epoch": 1.1477536462023306, + "grad_norm": 0.001139773172326386, + "learning_rate": 0.10940867294461679, + "loss": 0.1194, + "num_input_tokens_seen": 30464928, + "step": 23490 + }, + { + "epoch": 1.147997947866025, + "grad_norm": 0.0008952537318691611, + "learning_rate": 0.10935196899030565, + "loss": 0.1035, + "num_input_tokens_seen": 30470976, + "step": 23495 + }, + { + "epoch": 1.1482422495297193, + "grad_norm": 0.0014143313746899366, + "learning_rate": 0.10929527130443177, + "loss": 0.0734, + "num_input_tokens_seen": 30477728, + "step": 23500 + }, + { + "epoch": 1.1484865511934137, + "grad_norm": 0.00045097569818608463, + "learning_rate": 0.1092385798957385, + "loss": 0.1011, + "num_input_tokens_seen": 30483776, + "step": 23505 + }, + { + "epoch": 1.148730852857108, + "grad_norm": 0.0013856363948434591, + "learning_rate": 0.10918189477296848, + "loss": 0.0869, + "num_input_tokens_seen": 30490144, + "step": 23510 + }, + { + "epoch": 1.1489751545208022, + "grad_norm": 0.0013580137165263295, + "learning_rate": 0.1091252159448633, + "loss": 0.0928, + "num_input_tokens_seen": 30496640, + "step": 23515 + }, + { + "epoch": 1.1492194561844966, + "grad_norm": 0.0007652966887690127, + "learning_rate": 0.10906854342016345, + "loss": 0.0897, + "num_input_tokens_seen": 30502944, + "step": 23520 + }, + { + "epoch": 1.149463757848191, + "grad_norm": 0.0011187723139300942, + "learning_rate": 0.10901187720760858, + "loss": 0.1142, + "num_input_tokens_seen": 30509088, + "step": 23525 + }, + { + "epoch": 1.1497080595118854, + "grad_norm": 0.0015025229658931494, + "learning_rate": 0.10895521731593734, + "loss": 0.1124, + "num_input_tokens_seen": 30515264, + "step": 23530 + }, + { + "epoch": 1.1499523611755795, + "grad_norm": 0.0013136774068698287, + "learning_rate": 0.10889856375388733, + "loss": 0.1009, + "num_input_tokens_seen": 30521440, + "step": 23535 + }, + { + "epoch": 1.150196662839274, + "grad_norm": 0.0015981068136170506, + "learning_rate": 0.1088419165301954, + "loss": 0.0945, + "num_input_tokens_seen": 30528000, + "step": 23540 + }, + { + "epoch": 1.1504409645029683, + "grad_norm": 0.0009201017674058676, + "learning_rate": 0.1087852756535971, + "loss": 0.0917, + "num_input_tokens_seen": 30534272, + "step": 23545 + }, + { + "epoch": 1.1506852661666627, + "grad_norm": 0.0015229912241920829, + "learning_rate": 0.10872864113282725, + "loss": 0.0894, + "num_input_tokens_seen": 30540384, + "step": 23550 + }, + { + "epoch": 1.150929567830357, + "grad_norm": 0.000792122446000576, + "learning_rate": 0.10867201297661958, + "loss": 0.0921, + "num_input_tokens_seen": 30546944, + "step": 23555 + }, + { + "epoch": 1.1511738694940512, + "grad_norm": 0.0009434618405066431, + "learning_rate": 0.10861539119370689, + "loss": 0.0789, + "num_input_tokens_seen": 30553888, + "step": 23560 + }, + { + "epoch": 1.1514181711577456, + "grad_norm": 0.0016932040452957153, + "learning_rate": 0.10855877579282096, + "loss": 0.0923, + "num_input_tokens_seen": 30560736, + "step": 23565 + }, + { + "epoch": 1.15166247282144, + "grad_norm": 0.001518811797723174, + "learning_rate": 0.10850216678269252, + "loss": 0.0956, + "num_input_tokens_seen": 30567264, + "step": 23570 + }, + { + "epoch": 1.1519067744851341, + "grad_norm": 0.0012218249030411243, + "learning_rate": 0.10844556417205146, + "loss": 0.0909, + "num_input_tokens_seen": 30573536, + "step": 23575 + }, + { + "epoch": 1.1521510761488285, + "grad_norm": 0.0025090384297072887, + "learning_rate": 0.10838896796962669, + "loss": 0.1021, + "num_input_tokens_seen": 30579776, + "step": 23580 + }, + { + "epoch": 1.1523953778125229, + "grad_norm": 0.0008992078364826739, + "learning_rate": 0.1083323781841459, + "loss": 0.0963, + "num_input_tokens_seen": 30586080, + "step": 23585 + }, + { + "epoch": 1.1526396794762173, + "grad_norm": 0.0014047937002032995, + "learning_rate": 0.10827579482433607, + "loss": 0.0979, + "num_input_tokens_seen": 30592480, + "step": 23590 + }, + { + "epoch": 1.1528839811399116, + "grad_norm": 0.0011052173795178533, + "learning_rate": 0.10821921789892304, + "loss": 0.0894, + "num_input_tokens_seen": 30598912, + "step": 23595 + }, + { + "epoch": 1.1531282828036058, + "grad_norm": 0.0008847339195199311, + "learning_rate": 0.10816264741663158, + "loss": 0.1152, + "num_input_tokens_seen": 30605344, + "step": 23600 + }, + { + "epoch": 1.1531282828036058, + "eval_loss": 0.09443467110395432, + "eval_runtime": 402.3019, + "eval_samples_per_second": 90.442, + "eval_steps_per_second": 22.612, + "num_input_tokens_seen": 30605344, + "step": 23600 + }, + { + "epoch": 1.1533725844673002, + "grad_norm": 0.0010621803812682629, + "learning_rate": 0.10810608338618573, + "loss": 0.0686, + "num_input_tokens_seen": 30612416, + "step": 23605 + }, + { + "epoch": 1.1536168861309946, + "grad_norm": 0.0013327214401215315, + "learning_rate": 0.10804952581630821, + "loss": 0.0965, + "num_input_tokens_seen": 30618464, + "step": 23610 + }, + { + "epoch": 1.153861187794689, + "grad_norm": 0.0009897592244669795, + "learning_rate": 0.10799297471572102, + "loss": 0.0702, + "num_input_tokens_seen": 30624704, + "step": 23615 + }, + { + "epoch": 1.154105489458383, + "grad_norm": 0.0011910221073776484, + "learning_rate": 0.10793643009314507, + "loss": 0.0851, + "num_input_tokens_seen": 30631008, + "step": 23620 + }, + { + "epoch": 1.1543497911220775, + "grad_norm": 0.0008548352052457631, + "learning_rate": 0.10787989195730015, + "loss": 0.0678, + "num_input_tokens_seen": 30637792, + "step": 23625 + }, + { + "epoch": 1.1545940927857719, + "grad_norm": 0.0005340336356312037, + "learning_rate": 0.10782336031690525, + "loss": 0.0955, + "num_input_tokens_seen": 30644096, + "step": 23630 + }, + { + "epoch": 1.1548383944494662, + "grad_norm": 0.000664780440274626, + "learning_rate": 0.10776683518067821, + "loss": 0.0812, + "num_input_tokens_seen": 30650848, + "step": 23635 + }, + { + "epoch": 1.1550826961131606, + "grad_norm": 0.0019160029478371143, + "learning_rate": 0.10771031655733587, + "loss": 0.0823, + "num_input_tokens_seen": 30657664, + "step": 23640 + }, + { + "epoch": 1.1553269977768548, + "grad_norm": 0.0009486329508945346, + "learning_rate": 0.10765380445559422, + "loss": 0.0834, + "num_input_tokens_seen": 30664640, + "step": 23645 + }, + { + "epoch": 1.1555712994405491, + "grad_norm": 0.0011394219473004341, + "learning_rate": 0.10759729888416801, + "loss": 0.0756, + "num_input_tokens_seen": 30671488, + "step": 23650 + }, + { + "epoch": 1.1558156011042435, + "grad_norm": 0.0009731571190059185, + "learning_rate": 0.10754079985177119, + "loss": 0.0896, + "num_input_tokens_seen": 30677952, + "step": 23655 + }, + { + "epoch": 1.156059902767938, + "grad_norm": 0.001346671488136053, + "learning_rate": 0.10748430736711667, + "loss": 0.0946, + "num_input_tokens_seen": 30684384, + "step": 23660 + }, + { + "epoch": 1.156304204431632, + "grad_norm": 0.0019759631250053644, + "learning_rate": 0.10742782143891623, + "loss": 0.1007, + "num_input_tokens_seen": 30690848, + "step": 23665 + }, + { + "epoch": 1.1565485060953264, + "grad_norm": 0.000944876519497484, + "learning_rate": 0.10737134207588069, + "loss": 0.0806, + "num_input_tokens_seen": 30697728, + "step": 23670 + }, + { + "epoch": 1.1567928077590208, + "grad_norm": 0.0006841050926595926, + "learning_rate": 0.10731486928671992, + "loss": 0.0784, + "num_input_tokens_seen": 30704160, + "step": 23675 + }, + { + "epoch": 1.1570371094227152, + "grad_norm": 0.001622232492081821, + "learning_rate": 0.10725840308014269, + "loss": 0.1115, + "num_input_tokens_seen": 30710752, + "step": 23680 + }, + { + "epoch": 1.1572814110864096, + "grad_norm": 0.0013301329454407096, + "learning_rate": 0.10720194346485688, + "loss": 0.1001, + "num_input_tokens_seen": 30717568, + "step": 23685 + }, + { + "epoch": 1.1575257127501037, + "grad_norm": 0.0013941696379333735, + "learning_rate": 0.10714549044956918, + "loss": 0.0872, + "num_input_tokens_seen": 30723968, + "step": 23690 + }, + { + "epoch": 1.1577700144137981, + "grad_norm": 0.0012136942241340876, + "learning_rate": 0.10708904404298542, + "loss": 0.0841, + "num_input_tokens_seen": 30730592, + "step": 23695 + }, + { + "epoch": 1.1580143160774925, + "grad_norm": 0.0010109238792210817, + "learning_rate": 0.1070326042538103, + "loss": 0.1067, + "num_input_tokens_seen": 30737024, + "step": 23700 + }, + { + "epoch": 1.1582586177411869, + "grad_norm": 0.0011632655514404178, + "learning_rate": 0.10697617109074758, + "loss": 0.1142, + "num_input_tokens_seen": 30743136, + "step": 23705 + }, + { + "epoch": 1.158502919404881, + "grad_norm": 0.0010603271657600999, + "learning_rate": 0.10691974456249999, + "loss": 0.0878, + "num_input_tokens_seen": 30749280, + "step": 23710 + }, + { + "epoch": 1.1587472210685754, + "grad_norm": 0.0007519408245570958, + "learning_rate": 0.10686332467776909, + "loss": 0.0748, + "num_input_tokens_seen": 30755776, + "step": 23715 + }, + { + "epoch": 1.1589915227322698, + "grad_norm": 0.0012752395123243332, + "learning_rate": 0.10680691144525563, + "loss": 0.1098, + "num_input_tokens_seen": 30762368, + "step": 23720 + }, + { + "epoch": 1.1592358243959642, + "grad_norm": 0.0016394193517044187, + "learning_rate": 0.10675050487365928, + "loss": 0.0932, + "num_input_tokens_seen": 30769024, + "step": 23725 + }, + { + "epoch": 1.1594801260596586, + "grad_norm": 0.0016187210567295551, + "learning_rate": 0.10669410497167851, + "loss": 0.1087, + "num_input_tokens_seen": 30775456, + "step": 23730 + }, + { + "epoch": 1.1597244277233527, + "grad_norm": 0.0008881499525159597, + "learning_rate": 0.10663771174801102, + "loss": 0.0994, + "num_input_tokens_seen": 30781824, + "step": 23735 + }, + { + "epoch": 1.159968729387047, + "grad_norm": 0.0011099884286522865, + "learning_rate": 0.10658132521135329, + "loss": 0.083, + "num_input_tokens_seen": 30788576, + "step": 23740 + }, + { + "epoch": 1.1602130310507415, + "grad_norm": 0.0010162538383156061, + "learning_rate": 0.10652494537040084, + "loss": 0.0874, + "num_input_tokens_seen": 30795008, + "step": 23745 + }, + { + "epoch": 1.1604573327144359, + "grad_norm": 0.0012703813845291734, + "learning_rate": 0.1064685722338482, + "loss": 0.1112, + "num_input_tokens_seen": 30801664, + "step": 23750 + }, + { + "epoch": 1.16070163437813, + "grad_norm": 0.0013635392533615232, + "learning_rate": 0.10641220581038871, + "loss": 0.0761, + "num_input_tokens_seen": 30808256, + "step": 23755 + }, + { + "epoch": 1.1609459360418244, + "grad_norm": 0.001299510826356709, + "learning_rate": 0.10635584610871483, + "loss": 0.1092, + "num_input_tokens_seen": 30815072, + "step": 23760 + }, + { + "epoch": 1.1611902377055188, + "grad_norm": 0.0012526905629783869, + "learning_rate": 0.10629949313751803, + "loss": 0.1107, + "num_input_tokens_seen": 30821632, + "step": 23765 + }, + { + "epoch": 1.1614345393692131, + "grad_norm": 0.001192797557450831, + "learning_rate": 0.10624314690548849, + "loss": 0.0888, + "num_input_tokens_seen": 30828032, + "step": 23770 + }, + { + "epoch": 1.1616788410329075, + "grad_norm": 0.0008715243311598897, + "learning_rate": 0.1061868074213156, + "loss": 0.0744, + "num_input_tokens_seen": 30835008, + "step": 23775 + }, + { + "epoch": 1.1619231426966017, + "grad_norm": 0.0006626482936553657, + "learning_rate": 0.10613047469368765, + "loss": 0.0953, + "num_input_tokens_seen": 30841664, + "step": 23780 + }, + { + "epoch": 1.162167444360296, + "grad_norm": 0.0012190857669338584, + "learning_rate": 0.10607414873129171, + "loss": 0.0727, + "num_input_tokens_seen": 30847616, + "step": 23785 + }, + { + "epoch": 1.1624117460239904, + "grad_norm": 0.0009940799791365862, + "learning_rate": 0.10601782954281413, + "loss": 0.0845, + "num_input_tokens_seen": 30854272, + "step": 23790 + }, + { + "epoch": 1.1626560476876848, + "grad_norm": 0.0011105991434305906, + "learning_rate": 0.1059615171369399, + "loss": 0.108, + "num_input_tokens_seen": 30860960, + "step": 23795 + }, + { + "epoch": 1.162900349351379, + "grad_norm": 0.0010812492109835148, + "learning_rate": 0.10590521152235312, + "loss": 0.0853, + "num_input_tokens_seen": 30867648, + "step": 23800 + }, + { + "epoch": 1.162900349351379, + "eval_loss": 0.091725192964077, + "eval_runtime": 401.889, + "eval_samples_per_second": 90.535, + "eval_steps_per_second": 22.636, + "num_input_tokens_seen": 30867648, + "step": 23800 + }, + { + "epoch": 1.1631446510150734, + "grad_norm": 0.0009772691410034895, + "learning_rate": 0.1058489127077369, + "loss": 0.099, + "num_input_tokens_seen": 30873696, + "step": 23805 + }, + { + "epoch": 1.1633889526787677, + "grad_norm": 0.0009635406895540655, + "learning_rate": 0.1057926207017732, + "loss": 0.1063, + "num_input_tokens_seen": 30880064, + "step": 23810 + }, + { + "epoch": 1.1636332543424621, + "grad_norm": 0.0005893849884159863, + "learning_rate": 0.10573633551314285, + "loss": 0.0682, + "num_input_tokens_seen": 30886816, + "step": 23815 + }, + { + "epoch": 1.1638775560061565, + "grad_norm": 0.0008112158393487334, + "learning_rate": 0.1056800571505259, + "loss": 0.0739, + "num_input_tokens_seen": 30893760, + "step": 23820 + }, + { + "epoch": 1.1641218576698507, + "grad_norm": 0.0008587480988353491, + "learning_rate": 0.10562378562260105, + "loss": 0.0953, + "num_input_tokens_seen": 30900384, + "step": 23825 + }, + { + "epoch": 1.164366159333545, + "grad_norm": 0.0011753750732168555, + "learning_rate": 0.10556752093804615, + "loss": 0.0981, + "num_input_tokens_seen": 30907488, + "step": 23830 + }, + { + "epoch": 1.1646104609972394, + "grad_norm": 0.0009785124566406012, + "learning_rate": 0.10551126310553786, + "loss": 0.0763, + "num_input_tokens_seen": 30914112, + "step": 23835 + }, + { + "epoch": 1.1648547626609338, + "grad_norm": 0.0014366719406098127, + "learning_rate": 0.10545501213375187, + "loss": 0.0935, + "num_input_tokens_seen": 30920288, + "step": 23840 + }, + { + "epoch": 1.165099064324628, + "grad_norm": 0.0010301036527380347, + "learning_rate": 0.10539876803136287, + "loss": 0.0805, + "num_input_tokens_seen": 30927168, + "step": 23845 + }, + { + "epoch": 1.1653433659883223, + "grad_norm": 0.0014814691385254264, + "learning_rate": 0.10534253080704428, + "loss": 0.0959, + "num_input_tokens_seen": 30933600, + "step": 23850 + }, + { + "epoch": 1.1655876676520167, + "grad_norm": 0.0015058700228109956, + "learning_rate": 0.10528630046946862, + "loss": 0.0943, + "num_input_tokens_seen": 30940224, + "step": 23855 + }, + { + "epoch": 1.165831969315711, + "grad_norm": 0.0011235146084800363, + "learning_rate": 0.1052300770273074, + "loss": 0.0731, + "num_input_tokens_seen": 30946912, + "step": 23860 + }, + { + "epoch": 1.1660762709794055, + "grad_norm": 0.0009912211680784822, + "learning_rate": 0.10517386048923086, + "loss": 0.106, + "num_input_tokens_seen": 30953376, + "step": 23865 + }, + { + "epoch": 1.1663205726430996, + "grad_norm": 0.001013339962810278, + "learning_rate": 0.10511765086390841, + "loss": 0.0669, + "num_input_tokens_seen": 30959808, + "step": 23870 + }, + { + "epoch": 1.166564874306794, + "grad_norm": 0.001256113755516708, + "learning_rate": 0.10506144816000816, + "loss": 0.1045, + "num_input_tokens_seen": 30966304, + "step": 23875 + }, + { + "epoch": 1.1668091759704884, + "grad_norm": 0.001026311656460166, + "learning_rate": 0.10500525238619736, + "loss": 0.096, + "num_input_tokens_seen": 30972576, + "step": 23880 + }, + { + "epoch": 1.1670534776341828, + "grad_norm": 0.0014752852730453014, + "learning_rate": 0.10494906355114209, + "loss": 0.1023, + "num_input_tokens_seen": 30978976, + "step": 23885 + }, + { + "epoch": 1.167297779297877, + "grad_norm": 0.001568425097502768, + "learning_rate": 0.10489288166350737, + "loss": 0.086, + "num_input_tokens_seen": 30985568, + "step": 23890 + }, + { + "epoch": 1.1675420809615713, + "grad_norm": 0.0011794513557106256, + "learning_rate": 0.10483670673195711, + "loss": 0.07, + "num_input_tokens_seen": 30992224, + "step": 23895 + }, + { + "epoch": 1.1677863826252657, + "grad_norm": 0.0007672258070670068, + "learning_rate": 0.10478053876515431, + "loss": 0.078, + "num_input_tokens_seen": 30999008, + "step": 23900 + }, + { + "epoch": 1.16803068428896, + "grad_norm": 0.0011259311577305198, + "learning_rate": 0.10472437777176061, + "loss": 0.0771, + "num_input_tokens_seen": 31005408, + "step": 23905 + }, + { + "epoch": 1.1682749859526544, + "grad_norm": 0.0011640188749879599, + "learning_rate": 0.1046682237604369, + "loss": 0.1109, + "num_input_tokens_seen": 31011520, + "step": 23910 + }, + { + "epoch": 1.1685192876163486, + "grad_norm": 0.0011094182264059782, + "learning_rate": 0.1046120767398427, + "loss": 0.1066, + "num_input_tokens_seen": 31017792, + "step": 23915 + }, + { + "epoch": 1.168763589280043, + "grad_norm": 0.0007111740997061133, + "learning_rate": 0.10455593671863667, + "loss": 0.076, + "num_input_tokens_seen": 31024352, + "step": 23920 + }, + { + "epoch": 1.1690078909437374, + "grad_norm": 0.0011380012147128582, + "learning_rate": 0.1044998037054763, + "loss": 0.0974, + "num_input_tokens_seen": 31030816, + "step": 23925 + }, + { + "epoch": 1.1692521926074317, + "grad_norm": 0.0008126604370772839, + "learning_rate": 0.10444367770901794, + "loss": 0.0641, + "num_input_tokens_seen": 31036736, + "step": 23930 + }, + { + "epoch": 1.169496494271126, + "grad_norm": 0.0012198706390336156, + "learning_rate": 0.10438755873791698, + "loss": 0.0872, + "num_input_tokens_seen": 31043392, + "step": 23935 + }, + { + "epoch": 1.1697407959348203, + "grad_norm": 0.00088619499001652, + "learning_rate": 0.10433144680082775, + "loss": 0.1016, + "num_input_tokens_seen": 31049888, + "step": 23940 + }, + { + "epoch": 1.1699850975985147, + "grad_norm": 0.0007288389024324715, + "learning_rate": 0.10427534190640322, + "loss": 0.0846, + "num_input_tokens_seen": 31056736, + "step": 23945 + }, + { + "epoch": 1.170229399262209, + "grad_norm": 0.0016335879918187857, + "learning_rate": 0.10421924406329568, + "loss": 0.1074, + "num_input_tokens_seen": 31063424, + "step": 23950 + }, + { + "epoch": 1.1704737009259034, + "grad_norm": 0.0011474454076960683, + "learning_rate": 0.10416315328015598, + "loss": 0.0759, + "num_input_tokens_seen": 31069760, + "step": 23955 + }, + { + "epoch": 1.1707180025895976, + "grad_norm": 0.0011178847635164857, + "learning_rate": 0.10410706956563402, + "loss": 0.0842, + "num_input_tokens_seen": 31076352, + "step": 23960 + }, + { + "epoch": 1.170962304253292, + "grad_norm": 0.0009841988794505596, + "learning_rate": 0.10405099292837874, + "loss": 0.0971, + "num_input_tokens_seen": 31082560, + "step": 23965 + }, + { + "epoch": 1.1712066059169863, + "grad_norm": 0.0011507375165820122, + "learning_rate": 0.10399492337703771, + "loss": 0.0731, + "num_input_tokens_seen": 31089152, + "step": 23970 + }, + { + "epoch": 1.1714509075806807, + "grad_norm": 0.0009373925859108567, + "learning_rate": 0.10393886092025764, + "loss": 0.0737, + "num_input_tokens_seen": 31095744, + "step": 23975 + }, + { + "epoch": 1.1716952092443749, + "grad_norm": 0.0009606504463590682, + "learning_rate": 0.10388280556668412, + "loss": 0.0935, + "num_input_tokens_seen": 31102272, + "step": 23980 + }, + { + "epoch": 1.1719395109080692, + "grad_norm": 0.0009738794760778546, + "learning_rate": 0.10382675732496145, + "loss": 0.099, + "num_input_tokens_seen": 31108832, + "step": 23985 + }, + { + "epoch": 1.1721838125717636, + "grad_norm": 0.0021371867042034864, + "learning_rate": 0.10377071620373311, + "loss": 0.0911, + "num_input_tokens_seen": 31114912, + "step": 23990 + }, + { + "epoch": 1.172428114235458, + "grad_norm": 0.0014040411915630102, + "learning_rate": 0.10371468221164128, + "loss": 0.0779, + "num_input_tokens_seen": 31121280, + "step": 23995 + }, + { + "epoch": 1.1726724158991524, + "grad_norm": 0.0010924929520115256, + "learning_rate": 0.10365865535732706, + "loss": 0.0941, + "num_input_tokens_seen": 31127744, + "step": 24000 + }, + { + "epoch": 1.1726724158991524, + "eval_loss": 0.09170150756835938, + "eval_runtime": 402.4015, + "eval_samples_per_second": 90.42, + "eval_steps_per_second": 22.607, + "num_input_tokens_seen": 31127744, + "step": 24000 + }, + { + "epoch": 1.1729167175628465, + "grad_norm": 0.0011791607830673456, + "learning_rate": 0.10360263564943062, + "loss": 0.0983, + "num_input_tokens_seen": 31134272, + "step": 24005 + }, + { + "epoch": 1.173161019226541, + "grad_norm": 0.0014124218141660094, + "learning_rate": 0.10354662309659075, + "loss": 0.0998, + "num_input_tokens_seen": 31140640, + "step": 24010 + }, + { + "epoch": 1.1734053208902353, + "grad_norm": 0.0012583944480866194, + "learning_rate": 0.10349061770744537, + "loss": 0.1068, + "num_input_tokens_seen": 31146368, + "step": 24015 + }, + { + "epoch": 1.1736496225539297, + "grad_norm": 0.000938179437071085, + "learning_rate": 0.10343461949063128, + "loss": 0.0879, + "num_input_tokens_seen": 31152480, + "step": 24020 + }, + { + "epoch": 1.1738939242176238, + "grad_norm": 0.0014682415639981627, + "learning_rate": 0.103378628454784, + "loss": 0.0902, + "num_input_tokens_seen": 31158784, + "step": 24025 + }, + { + "epoch": 1.1741382258813182, + "grad_norm": 0.0011140468996018171, + "learning_rate": 0.10332264460853811, + "loss": 0.0918, + "num_input_tokens_seen": 31165248, + "step": 24030 + }, + { + "epoch": 1.1743825275450126, + "grad_norm": 0.0010216737864539027, + "learning_rate": 0.10326666796052701, + "loss": 0.083, + "num_input_tokens_seen": 31171360, + "step": 24035 + }, + { + "epoch": 1.174626829208707, + "grad_norm": 0.0009753673220984638, + "learning_rate": 0.10321069851938296, + "loss": 0.0888, + "num_input_tokens_seen": 31177760, + "step": 24040 + }, + { + "epoch": 1.1748711308724014, + "grad_norm": 0.0021770326420664787, + "learning_rate": 0.10315473629373724, + "loss": 0.1112, + "num_input_tokens_seen": 31184160, + "step": 24045 + }, + { + "epoch": 1.1751154325360955, + "grad_norm": 0.000951842637732625, + "learning_rate": 0.10309878129221982, + "loss": 0.1048, + "num_input_tokens_seen": 31190944, + "step": 24050 + }, + { + "epoch": 1.17535973419979, + "grad_norm": 0.0010287126060575247, + "learning_rate": 0.10304283352345973, + "loss": 0.1176, + "num_input_tokens_seen": 31197440, + "step": 24055 + }, + { + "epoch": 1.1756040358634843, + "grad_norm": 0.0009578716126270592, + "learning_rate": 0.10298689299608486, + "loss": 0.0963, + "num_input_tokens_seen": 31204352, + "step": 24060 + }, + { + "epoch": 1.1758483375271787, + "grad_norm": 0.0013073625741526484, + "learning_rate": 0.10293095971872188, + "loss": 0.0794, + "num_input_tokens_seen": 31210752, + "step": 24065 + }, + { + "epoch": 1.1760926391908728, + "grad_norm": 0.0010995222255587578, + "learning_rate": 0.10287503369999645, + "loss": 0.0894, + "num_input_tokens_seen": 31217792, + "step": 24070 + }, + { + "epoch": 1.1763369408545672, + "grad_norm": 0.0014191775117069483, + "learning_rate": 0.10281911494853295, + "loss": 0.0806, + "num_input_tokens_seen": 31224320, + "step": 24075 + }, + { + "epoch": 1.1765812425182616, + "grad_norm": 0.0010905605740845203, + "learning_rate": 0.10276320347295485, + "loss": 0.0801, + "num_input_tokens_seen": 31230592, + "step": 24080 + }, + { + "epoch": 1.176825544181956, + "grad_norm": 0.001413712278008461, + "learning_rate": 0.10270729928188446, + "loss": 0.1067, + "num_input_tokens_seen": 31236416, + "step": 24085 + }, + { + "epoch": 1.1770698458456503, + "grad_norm": 0.0012503372272476554, + "learning_rate": 0.10265140238394276, + "loss": 0.1009, + "num_input_tokens_seen": 31242656, + "step": 24090 + }, + { + "epoch": 1.1773141475093445, + "grad_norm": 0.0014419788494706154, + "learning_rate": 0.10259551278774988, + "loss": 0.0799, + "num_input_tokens_seen": 31249280, + "step": 24095 + }, + { + "epoch": 1.1775584491730389, + "grad_norm": 0.0012613738654181361, + "learning_rate": 0.10253963050192462, + "loss": 0.0774, + "num_input_tokens_seen": 31255968, + "step": 24100 + }, + { + "epoch": 1.1778027508367332, + "grad_norm": 0.0007075080065988004, + "learning_rate": 0.10248375553508478, + "loss": 0.0706, + "num_input_tokens_seen": 31262208, + "step": 24105 + }, + { + "epoch": 1.1780470525004274, + "grad_norm": 0.0022465756628662348, + "learning_rate": 0.102427887895847, + "loss": 0.138, + "num_input_tokens_seen": 31268160, + "step": 24110 + }, + { + "epoch": 1.1782913541641218, + "grad_norm": 0.0011205562623217702, + "learning_rate": 0.10237202759282668, + "loss": 0.0767, + "num_input_tokens_seen": 31274368, + "step": 24115 + }, + { + "epoch": 1.1785356558278162, + "grad_norm": 0.0015615872107446194, + "learning_rate": 0.10231617463463821, + "loss": 0.0762, + "num_input_tokens_seen": 31280416, + "step": 24120 + }, + { + "epoch": 1.1787799574915105, + "grad_norm": 0.0009780417894944549, + "learning_rate": 0.10226032902989492, + "loss": 0.1137, + "num_input_tokens_seen": 31286656, + "step": 24125 + }, + { + "epoch": 1.179024259155205, + "grad_norm": 0.0018251813016831875, + "learning_rate": 0.10220449078720877, + "loss": 0.1043, + "num_input_tokens_seen": 31292736, + "step": 24130 + }, + { + "epoch": 1.179268560818899, + "grad_norm": 0.0013035794254392385, + "learning_rate": 0.1021486599151908, + "loss": 0.0986, + "num_input_tokens_seen": 31299008, + "step": 24135 + }, + { + "epoch": 1.1795128624825935, + "grad_norm": 0.0018444120651111007, + "learning_rate": 0.10209283642245084, + "loss": 0.0841, + "num_input_tokens_seen": 31305568, + "step": 24140 + }, + { + "epoch": 1.1797571641462878, + "grad_norm": 0.0007441001944243908, + "learning_rate": 0.10203702031759748, + "loss": 0.0703, + "num_input_tokens_seen": 31311936, + "step": 24145 + }, + { + "epoch": 1.1800014658099822, + "grad_norm": 0.0012367567978799343, + "learning_rate": 0.1019812116092384, + "loss": 0.0966, + "num_input_tokens_seen": 31318208, + "step": 24150 + }, + { + "epoch": 1.1802457674736764, + "grad_norm": 0.002065615728497505, + "learning_rate": 0.10192541030597986, + "loss": 0.0938, + "num_input_tokens_seen": 31324512, + "step": 24155 + }, + { + "epoch": 1.1804900691373708, + "grad_norm": 0.0014458007644861937, + "learning_rate": 0.1018696164164272, + "loss": 0.0681, + "num_input_tokens_seen": 31331584, + "step": 24160 + }, + { + "epoch": 1.1807343708010651, + "grad_norm": 0.0011744268704205751, + "learning_rate": 0.10181382994918459, + "loss": 0.0766, + "num_input_tokens_seen": 31338016, + "step": 24165 + }, + { + "epoch": 1.1809786724647595, + "grad_norm": 0.00108047085814178, + "learning_rate": 0.10175805091285492, + "loss": 0.1011, + "num_input_tokens_seen": 31344640, + "step": 24170 + }, + { + "epoch": 1.181222974128454, + "grad_norm": 0.0014539766125380993, + "learning_rate": 0.10170227931603999, + "loss": 0.0808, + "num_input_tokens_seen": 31351040, + "step": 24175 + }, + { + "epoch": 1.181467275792148, + "grad_norm": 0.0007090603467077017, + "learning_rate": 0.10164651516734062, + "loss": 0.0619, + "num_input_tokens_seen": 31357664, + "step": 24180 + }, + { + "epoch": 1.1817115774558424, + "grad_norm": 0.0009503260371275246, + "learning_rate": 0.1015907584753562, + "loss": 0.0852, + "num_input_tokens_seen": 31364064, + "step": 24185 + }, + { + "epoch": 1.1819558791195368, + "grad_norm": 0.0007459775079041719, + "learning_rate": 0.10153500924868523, + "loss": 0.0923, + "num_input_tokens_seen": 31370592, + "step": 24190 + }, + { + "epoch": 1.1822001807832312, + "grad_norm": 0.0014360620407387614, + "learning_rate": 0.10147926749592483, + "loss": 0.0772, + "num_input_tokens_seen": 31377120, + "step": 24195 + }, + { + "epoch": 1.1824444824469253, + "grad_norm": 0.0005053716595284641, + "learning_rate": 0.10142353322567112, + "loss": 0.0994, + "num_input_tokens_seen": 31383392, + "step": 24200 + }, + { + "epoch": 1.1824444824469253, + "eval_loss": 0.09184528887271881, + "eval_runtime": 402.5174, + "eval_samples_per_second": 90.394, + "eval_steps_per_second": 22.6, + "num_input_tokens_seen": 31383392, + "step": 24200 + }, + { + "epoch": 1.1826887841106197, + "grad_norm": 0.0012912340462207794, + "learning_rate": 0.1013678064465191, + "loss": 0.0691, + "num_input_tokens_seen": 31390304, + "step": 24205 + }, + { + "epoch": 1.182933085774314, + "grad_norm": 0.0010797976283356547, + "learning_rate": 0.10131208716706244, + "loss": 0.096, + "num_input_tokens_seen": 31396768, + "step": 24210 + }, + { + "epoch": 1.1831773874380085, + "grad_norm": 0.000915258249733597, + "learning_rate": 0.10125637539589379, + "loss": 0.071, + "num_input_tokens_seen": 31403232, + "step": 24215 + }, + { + "epoch": 1.1834216891017029, + "grad_norm": 0.001093352329917252, + "learning_rate": 0.10120067114160464, + "loss": 0.071, + "num_input_tokens_seen": 31410368, + "step": 24220 + }, + { + "epoch": 1.183665990765397, + "grad_norm": 0.0012661871733143926, + "learning_rate": 0.10114497441278517, + "loss": 0.0997, + "num_input_tokens_seen": 31416768, + "step": 24225 + }, + { + "epoch": 1.1839102924290914, + "grad_norm": 0.0016350619262084365, + "learning_rate": 0.10108928521802468, + "loss": 0.0812, + "num_input_tokens_seen": 31422976, + "step": 24230 + }, + { + "epoch": 1.1841545940927858, + "grad_norm": 0.0016345275798812509, + "learning_rate": 0.101033603565911, + "loss": 0.0878, + "num_input_tokens_seen": 31429600, + "step": 24235 + }, + { + "epoch": 1.1843988957564802, + "grad_norm": 0.0015674714231863618, + "learning_rate": 0.10097792946503102, + "loss": 0.1063, + "num_input_tokens_seen": 31436256, + "step": 24240 + }, + { + "epoch": 1.1846431974201743, + "grad_norm": 0.0018953140825033188, + "learning_rate": 0.10092226292397039, + "loss": 0.1046, + "num_input_tokens_seen": 31442304, + "step": 24245 + }, + { + "epoch": 1.1848874990838687, + "grad_norm": 0.002342848340049386, + "learning_rate": 0.10086660395131354, + "loss": 0.1002, + "num_input_tokens_seen": 31448832, + "step": 24250 + }, + { + "epoch": 1.185131800747563, + "grad_norm": 0.0014210615772753954, + "learning_rate": 0.10081095255564385, + "loss": 0.111, + "num_input_tokens_seen": 31455616, + "step": 24255 + }, + { + "epoch": 1.1853761024112575, + "grad_norm": 0.001204928383231163, + "learning_rate": 0.10075530874554335, + "loss": 0.087, + "num_input_tokens_seen": 31462016, + "step": 24260 + }, + { + "epoch": 1.1856204040749518, + "grad_norm": 0.0011940719559788704, + "learning_rate": 0.10069967252959311, + "loss": 0.0828, + "num_input_tokens_seen": 31468288, + "step": 24265 + }, + { + "epoch": 1.185864705738646, + "grad_norm": 0.0007091877632774413, + "learning_rate": 0.10064404391637297, + "loss": 0.0791, + "num_input_tokens_seen": 31474720, + "step": 24270 + }, + { + "epoch": 1.1861090074023404, + "grad_norm": 0.0016542894300073385, + "learning_rate": 0.10058842291446145, + "loss": 0.0969, + "num_input_tokens_seen": 31481376, + "step": 24275 + }, + { + "epoch": 1.1863533090660348, + "grad_norm": 0.0010810964740812778, + "learning_rate": 0.10053280953243608, + "loss": 0.0826, + "num_input_tokens_seen": 31487584, + "step": 24280 + }, + { + "epoch": 1.1865976107297291, + "grad_norm": 0.0014508579624816775, + "learning_rate": 0.10047720377887315, + "loss": 0.0968, + "num_input_tokens_seen": 31493760, + "step": 24285 + }, + { + "epoch": 1.1868419123934233, + "grad_norm": 0.0008820039220154285, + "learning_rate": 0.10042160566234767, + "loss": 0.0951, + "num_input_tokens_seen": 31499936, + "step": 24290 + }, + { + "epoch": 1.1870862140571177, + "grad_norm": 0.0012958049774169922, + "learning_rate": 0.10036601519143372, + "loss": 0.0912, + "num_input_tokens_seen": 31505984, + "step": 24295 + }, + { + "epoch": 1.187330515720812, + "grad_norm": 0.0014083333080634475, + "learning_rate": 0.1003104323747039, + "loss": 0.093, + "num_input_tokens_seen": 31512640, + "step": 24300 + }, + { + "epoch": 1.1875748173845064, + "grad_norm": 0.0010803603800013661, + "learning_rate": 0.10025485722072984, + "loss": 0.0854, + "num_input_tokens_seen": 31518816, + "step": 24305 + }, + { + "epoch": 1.1878191190482008, + "grad_norm": 0.001464216155000031, + "learning_rate": 0.10019928973808201, + "loss": 0.1075, + "num_input_tokens_seen": 31525280, + "step": 24310 + }, + { + "epoch": 1.188063420711895, + "grad_norm": 0.0014994036173447967, + "learning_rate": 0.10014372993532945, + "loss": 0.093, + "num_input_tokens_seen": 31531584, + "step": 24315 + }, + { + "epoch": 1.1883077223755893, + "grad_norm": 0.0013139928923919797, + "learning_rate": 0.1000881778210403, + "loss": 0.1058, + "num_input_tokens_seen": 31538048, + "step": 24320 + }, + { + "epoch": 1.1885520240392837, + "grad_norm": 0.0008505535079166293, + "learning_rate": 0.10003263340378142, + "loss": 0.094, + "num_input_tokens_seen": 31544800, + "step": 24325 + }, + { + "epoch": 1.188796325702978, + "grad_norm": 0.000685251725371927, + "learning_rate": 0.09997709669211834, + "loss": 0.0685, + "num_input_tokens_seen": 31551680, + "step": 24330 + }, + { + "epoch": 1.1890406273666723, + "grad_norm": 0.0010277159744873643, + "learning_rate": 0.0999215676946156, + "loss": 0.08, + "num_input_tokens_seen": 31558240, + "step": 24335 + }, + { + "epoch": 1.1892849290303666, + "grad_norm": 0.0007594769704155624, + "learning_rate": 0.0998660464198364, + "loss": 0.0946, + "num_input_tokens_seen": 31564608, + "step": 24340 + }, + { + "epoch": 1.189529230694061, + "grad_norm": 0.0010806324426084757, + "learning_rate": 0.09981053287634288, + "loss": 0.1098, + "num_input_tokens_seen": 31570880, + "step": 24345 + }, + { + "epoch": 1.1897735323577554, + "grad_norm": 0.0016616192879155278, + "learning_rate": 0.09975502707269596, + "loss": 0.088, + "num_input_tokens_seen": 31577664, + "step": 24350 + }, + { + "epoch": 1.1900178340214498, + "grad_norm": 0.0007256076205521822, + "learning_rate": 0.09969952901745524, + "loss": 0.0829, + "num_input_tokens_seen": 31584128, + "step": 24355 + }, + { + "epoch": 1.190262135685144, + "grad_norm": 0.0005287114763632417, + "learning_rate": 0.09964403871917925, + "loss": 0.0799, + "num_input_tokens_seen": 31590592, + "step": 24360 + }, + { + "epoch": 1.1905064373488383, + "grad_norm": 0.0014495333889499307, + "learning_rate": 0.09958855618642536, + "loss": 0.1018, + "num_input_tokens_seen": 31596864, + "step": 24365 + }, + { + "epoch": 1.1907507390125327, + "grad_norm": 0.0016420881729573011, + "learning_rate": 0.09953308142774955, + "loss": 0.0945, + "num_input_tokens_seen": 31603040, + "step": 24370 + }, + { + "epoch": 1.190995040676227, + "grad_norm": 0.0008511843043379486, + "learning_rate": 0.09947761445170686, + "loss": 0.0709, + "num_input_tokens_seen": 31609248, + "step": 24375 + }, + { + "epoch": 1.1912393423399212, + "grad_norm": 0.0007924340898171067, + "learning_rate": 0.09942215526685086, + "loss": 0.0959, + "num_input_tokens_seen": 31615648, + "step": 24380 + }, + { + "epoch": 1.1914836440036156, + "grad_norm": 0.0014697861624881625, + "learning_rate": 0.09936670388173414, + "loss": 0.11, + "num_input_tokens_seen": 31621600, + "step": 24385 + }, + { + "epoch": 1.19172794566731, + "grad_norm": 0.0007851478294469416, + "learning_rate": 0.09931126030490799, + "loss": 0.0948, + "num_input_tokens_seen": 31628000, + "step": 24390 + }, + { + "epoch": 1.1919722473310044, + "grad_norm": 0.0010441525373607874, + "learning_rate": 0.0992558245449225, + "loss": 0.0818, + "num_input_tokens_seen": 31634560, + "step": 24395 + }, + { + "epoch": 1.1922165489946988, + "grad_norm": 0.0013035680167376995, + "learning_rate": 0.09920039661032651, + "loss": 0.0724, + "num_input_tokens_seen": 31641056, + "step": 24400 + }, + { + "epoch": 1.1922165489946988, + "eval_loss": 0.09734250605106354, + "eval_runtime": 402.7552, + "eval_samples_per_second": 90.34, + "eval_steps_per_second": 22.587, + "num_input_tokens_seen": 31641056, + "step": 24400 + }, + { + "epoch": 1.192460850658393, + "grad_norm": 0.0016806444618850946, + "learning_rate": 0.09914497650966782, + "loss": 0.108, + "num_input_tokens_seen": 31648064, + "step": 24405 + }, + { + "epoch": 1.1927051523220873, + "grad_norm": 0.0014672859106212854, + "learning_rate": 0.09908956425149276, + "loss": 0.0657, + "num_input_tokens_seen": 31654528, + "step": 24410 + }, + { + "epoch": 1.1929494539857817, + "grad_norm": 0.0009551233379170299, + "learning_rate": 0.09903415984434677, + "loss": 0.0757, + "num_input_tokens_seen": 31661216, + "step": 24415 + }, + { + "epoch": 1.193193755649476, + "grad_norm": 0.0012440432328730822, + "learning_rate": 0.09897876329677373, + "loss": 0.1092, + "num_input_tokens_seen": 31667584, + "step": 24420 + }, + { + "epoch": 1.1934380573131702, + "grad_norm": 0.000700084085110575, + "learning_rate": 0.09892337461731658, + "loss": 0.1022, + "num_input_tokens_seen": 31673728, + "step": 24425 + }, + { + "epoch": 1.1936823589768646, + "grad_norm": 0.001104610739275813, + "learning_rate": 0.09886799381451693, + "loss": 0.101, + "num_input_tokens_seen": 31679872, + "step": 24430 + }, + { + "epoch": 1.193926660640559, + "grad_norm": 0.0012179926270619035, + "learning_rate": 0.09881262089691521, + "loss": 0.0799, + "num_input_tokens_seen": 31686176, + "step": 24435 + }, + { + "epoch": 1.1941709623042533, + "grad_norm": 0.0008927297894842923, + "learning_rate": 0.09875725587305059, + "loss": 0.0724, + "num_input_tokens_seen": 31692544, + "step": 24440 + }, + { + "epoch": 1.1944152639679477, + "grad_norm": 0.0011230278760194778, + "learning_rate": 0.09870189875146111, + "loss": 0.0886, + "num_input_tokens_seen": 31699136, + "step": 24445 + }, + { + "epoch": 1.1946595656316419, + "grad_norm": 0.0013442145427688956, + "learning_rate": 0.09864654954068346, + "loss": 0.0746, + "num_input_tokens_seen": 31705472, + "step": 24450 + }, + { + "epoch": 1.1949038672953363, + "grad_norm": 0.0008908587624318898, + "learning_rate": 0.09859120824925326, + "loss": 0.0866, + "num_input_tokens_seen": 31711712, + "step": 24455 + }, + { + "epoch": 1.1951481689590306, + "grad_norm": 0.0015566773945465684, + "learning_rate": 0.09853587488570474, + "loss": 0.1107, + "num_input_tokens_seen": 31718368, + "step": 24460 + }, + { + "epoch": 1.195392470622725, + "grad_norm": 0.0012331882026046515, + "learning_rate": 0.09848054945857107, + "loss": 0.075, + "num_input_tokens_seen": 31724576, + "step": 24465 + }, + { + "epoch": 1.1956367722864192, + "grad_norm": 0.0011916793882846832, + "learning_rate": 0.09842523197638416, + "loss": 0.1121, + "num_input_tokens_seen": 31731776, + "step": 24470 + }, + { + "epoch": 1.1958810739501136, + "grad_norm": 0.000961719430051744, + "learning_rate": 0.09836992244767452, + "loss": 0.0878, + "num_input_tokens_seen": 31738400, + "step": 24475 + }, + { + "epoch": 1.196125375613808, + "grad_norm": 0.0010529770515859127, + "learning_rate": 0.09831462088097168, + "loss": 0.0872, + "num_input_tokens_seen": 31744384, + "step": 24480 + }, + { + "epoch": 1.1963696772775023, + "grad_norm": 0.0021849186159670353, + "learning_rate": 0.09825932728480385, + "loss": 0.0923, + "num_input_tokens_seen": 31750816, + "step": 24485 + }, + { + "epoch": 1.1966139789411967, + "grad_norm": 0.0012017997214570642, + "learning_rate": 0.09820404166769794, + "loss": 0.0955, + "num_input_tokens_seen": 31757344, + "step": 24490 + }, + { + "epoch": 1.1968582806048909, + "grad_norm": 0.0008814377360977232, + "learning_rate": 0.09814876403817978, + "loss": 0.1028, + "num_input_tokens_seen": 31764096, + "step": 24495 + }, + { + "epoch": 1.1971025822685852, + "grad_norm": 0.0012472885427996516, + "learning_rate": 0.09809349440477376, + "loss": 0.0838, + "num_input_tokens_seen": 31770528, + "step": 24500 + }, + { + "epoch": 1.1973468839322796, + "grad_norm": 0.0015322621911764145, + "learning_rate": 0.09803823277600317, + "loss": 0.1066, + "num_input_tokens_seen": 31776576, + "step": 24505 + }, + { + "epoch": 1.197591185595974, + "grad_norm": 0.0009051284869201481, + "learning_rate": 0.09798297916039014, + "loss": 0.0827, + "num_input_tokens_seen": 31782848, + "step": 24510 + }, + { + "epoch": 1.1978354872596682, + "grad_norm": 0.0015590114053338766, + "learning_rate": 0.09792773356645534, + "loss": 0.0761, + "num_input_tokens_seen": 31789696, + "step": 24515 + }, + { + "epoch": 1.1980797889233625, + "grad_norm": 0.0008781833457760513, + "learning_rate": 0.09787249600271843, + "loss": 0.0778, + "num_input_tokens_seen": 31796512, + "step": 24520 + }, + { + "epoch": 1.198324090587057, + "grad_norm": 0.0011107869213446975, + "learning_rate": 0.09781726647769776, + "loss": 0.1, + "num_input_tokens_seen": 31802272, + "step": 24525 + }, + { + "epoch": 1.1985683922507513, + "grad_norm": 0.0013004557695239782, + "learning_rate": 0.0977620449999103, + "loss": 0.1019, + "num_input_tokens_seen": 31809312, + "step": 24530 + }, + { + "epoch": 1.1988126939144457, + "grad_norm": 0.0007358186994679272, + "learning_rate": 0.09770683157787204, + "loss": 0.0947, + "num_input_tokens_seen": 31815424, + "step": 24535 + }, + { + "epoch": 1.1990569955781398, + "grad_norm": 0.0010960487416014075, + "learning_rate": 0.09765162622009745, + "loss": 0.0857, + "num_input_tokens_seen": 31822048, + "step": 24540 + }, + { + "epoch": 1.1993012972418342, + "grad_norm": 0.0007904191734269261, + "learning_rate": 0.09759642893509995, + "loss": 0.075, + "num_input_tokens_seen": 31828544, + "step": 24545 + }, + { + "epoch": 1.1995455989055286, + "grad_norm": 0.0018255648901686072, + "learning_rate": 0.09754123973139169, + "loss": 0.0924, + "num_input_tokens_seen": 31835424, + "step": 24550 + }, + { + "epoch": 1.199789900569223, + "grad_norm": 0.0013108161510899663, + "learning_rate": 0.09748605861748345, + "loss": 0.1053, + "num_input_tokens_seen": 31841856, + "step": 24555 + }, + { + "epoch": 1.2000342022329171, + "grad_norm": 0.0016831073444336653, + "learning_rate": 0.0974308856018849, + "loss": 0.0884, + "num_input_tokens_seen": 31848032, + "step": 24560 + }, + { + "epoch": 1.2002785038966115, + "grad_norm": 0.0014192695962265134, + "learning_rate": 0.09737572069310449, + "loss": 0.0824, + "num_input_tokens_seen": 31854464, + "step": 24565 + }, + { + "epoch": 1.2005228055603059, + "grad_norm": 0.0015908522764220834, + "learning_rate": 0.09732056389964922, + "loss": 0.0907, + "num_input_tokens_seen": 31860992, + "step": 24570 + }, + { + "epoch": 1.2007671072240003, + "grad_norm": 0.0007567342254333198, + "learning_rate": 0.097265415230025, + "loss": 0.0827, + "num_input_tokens_seen": 31867904, + "step": 24575 + }, + { + "epoch": 1.2010114088876946, + "grad_norm": 0.0007275679963640869, + "learning_rate": 0.09721027469273648, + "loss": 0.0883, + "num_input_tokens_seen": 31874048, + "step": 24580 + }, + { + "epoch": 1.2012557105513888, + "grad_norm": 0.000850810669362545, + "learning_rate": 0.09715514229628695, + "loss": 0.089, + "num_input_tokens_seen": 31881952, + "step": 24585 + }, + { + "epoch": 1.2015000122150832, + "grad_norm": 0.0013596676290035248, + "learning_rate": 0.09710001804917864, + "loss": 0.0943, + "num_input_tokens_seen": 31888256, + "step": 24590 + }, + { + "epoch": 1.2017443138787776, + "grad_norm": 0.0011327668325975537, + "learning_rate": 0.09704490195991226, + "loss": 0.0902, + "num_input_tokens_seen": 31894624, + "step": 24595 + }, + { + "epoch": 1.201988615542472, + "grad_norm": 0.001996656646952033, + "learning_rate": 0.09698979403698753, + "loss": 0.098, + "num_input_tokens_seen": 31900960, + "step": 24600 + }, + { + "epoch": 1.201988615542472, + "eval_loss": 0.09195403754711151, + "eval_runtime": 401.8348, + "eval_samples_per_second": 90.547, + "eval_steps_per_second": 22.639, + "num_input_tokens_seen": 31900960, + "step": 24600 + }, + { + "epoch": 1.202232917206166, + "grad_norm": 0.001125733251683414, + "learning_rate": 0.0969346942889027, + "loss": 0.0903, + "num_input_tokens_seen": 31907584, + "step": 24605 + }, + { + "epoch": 1.2024772188698605, + "grad_norm": 0.001516639138571918, + "learning_rate": 0.09687960272415487, + "loss": 0.0939, + "num_input_tokens_seen": 31914208, + "step": 24610 + }, + { + "epoch": 1.2027215205335549, + "grad_norm": 0.0014186511980369687, + "learning_rate": 0.0968245193512399, + "loss": 0.1173, + "num_input_tokens_seen": 31920128, + "step": 24615 + }, + { + "epoch": 1.2029658221972492, + "grad_norm": 0.0010320020373910666, + "learning_rate": 0.09676944417865221, + "loss": 0.1035, + "num_input_tokens_seen": 31926688, + "step": 24620 + }, + { + "epoch": 1.2032101238609436, + "grad_norm": 0.0008020127424970269, + "learning_rate": 0.09671437721488517, + "loss": 0.0601, + "num_input_tokens_seen": 31933152, + "step": 24625 + }, + { + "epoch": 1.2034544255246378, + "grad_norm": 0.0018813635688275099, + "learning_rate": 0.09665931846843086, + "loss": 0.0627, + "num_input_tokens_seen": 31940576, + "step": 24630 + }, + { + "epoch": 1.2036987271883322, + "grad_norm": 0.0009080271702259779, + "learning_rate": 0.0966042679477799, + "loss": 0.1026, + "num_input_tokens_seen": 31947360, + "step": 24635 + }, + { + "epoch": 1.2039430288520265, + "grad_norm": 0.0011042613768950105, + "learning_rate": 0.09654922566142186, + "loss": 0.0727, + "num_input_tokens_seen": 31953760, + "step": 24640 + }, + { + "epoch": 1.2041873305157207, + "grad_norm": 0.001175863086245954, + "learning_rate": 0.09649419161784498, + "loss": 0.0872, + "num_input_tokens_seen": 31960064, + "step": 24645 + }, + { + "epoch": 1.204431632179415, + "grad_norm": 0.0012827578466385603, + "learning_rate": 0.09643916582553606, + "loss": 0.0905, + "num_input_tokens_seen": 31967168, + "step": 24650 + }, + { + "epoch": 1.2046759338431094, + "grad_norm": 0.0008569850469939411, + "learning_rate": 0.09638414829298093, + "loss": 0.1005, + "num_input_tokens_seen": 31974144, + "step": 24655 + }, + { + "epoch": 1.2049202355068038, + "grad_norm": 0.0011768144322559237, + "learning_rate": 0.09632913902866386, + "loss": 0.0652, + "num_input_tokens_seen": 31980672, + "step": 24660 + }, + { + "epoch": 1.2051645371704982, + "grad_norm": 0.0010728446068242192, + "learning_rate": 0.096274138041068, + "loss": 0.0946, + "num_input_tokens_seen": 31987040, + "step": 24665 + }, + { + "epoch": 1.2054088388341924, + "grad_norm": 0.0010761613957583904, + "learning_rate": 0.09621914533867527, + "loss": 0.068, + "num_input_tokens_seen": 31993760, + "step": 24670 + }, + { + "epoch": 1.2056531404978867, + "grad_norm": 0.00174901622813195, + "learning_rate": 0.09616416092996616, + "loss": 0.0866, + "num_input_tokens_seen": 32000288, + "step": 24675 + }, + { + "epoch": 1.2058974421615811, + "grad_norm": 0.0010076502803713083, + "learning_rate": 0.09610918482342, + "loss": 0.0799, + "num_input_tokens_seen": 32006592, + "step": 24680 + }, + { + "epoch": 1.2061417438252755, + "grad_norm": 0.0009841980645433068, + "learning_rate": 0.09605421702751478, + "loss": 0.0891, + "num_input_tokens_seen": 32012352, + "step": 24685 + }, + { + "epoch": 1.2063860454889697, + "grad_norm": 0.0009788644965738058, + "learning_rate": 0.09599925755072718, + "loss": 0.0707, + "num_input_tokens_seen": 32018880, + "step": 24690 + }, + { + "epoch": 1.206630347152664, + "grad_norm": 0.0008545861928723752, + "learning_rate": 0.09594430640153273, + "loss": 0.1038, + "num_input_tokens_seen": 32025184, + "step": 24695 + }, + { + "epoch": 1.2068746488163584, + "grad_norm": 0.0016311880899593234, + "learning_rate": 0.09588936358840547, + "loss": 0.0944, + "num_input_tokens_seen": 32031648, + "step": 24700 + }, + { + "epoch": 1.2071189504800528, + "grad_norm": 0.0016771513037383556, + "learning_rate": 0.09583442911981836, + "loss": 0.0841, + "num_input_tokens_seen": 32038432, + "step": 24705 + }, + { + "epoch": 1.2073632521437472, + "grad_norm": 0.001691321493126452, + "learning_rate": 0.09577950300424302, + "loss": 0.092, + "num_input_tokens_seen": 32044384, + "step": 24710 + }, + { + "epoch": 1.2076075538074413, + "grad_norm": 0.0014130984200164676, + "learning_rate": 0.09572458525014967, + "loss": 0.1096, + "num_input_tokens_seen": 32050688, + "step": 24715 + }, + { + "epoch": 1.2078518554711357, + "grad_norm": 0.0015288463328033686, + "learning_rate": 0.0956696758660073, + "loss": 0.1223, + "num_input_tokens_seen": 32057216, + "step": 24720 + }, + { + "epoch": 1.20809615713483, + "grad_norm": 0.0016805172199383378, + "learning_rate": 0.09561477486028373, + "loss": 0.1089, + "num_input_tokens_seen": 32063680, + "step": 24725 + }, + { + "epoch": 1.2083404587985245, + "grad_norm": 0.0014676552964374423, + "learning_rate": 0.09555988224144528, + "loss": 0.0966, + "num_input_tokens_seen": 32070080, + "step": 24730 + }, + { + "epoch": 1.2085847604622186, + "grad_norm": 0.001188886002637446, + "learning_rate": 0.09550499801795717, + "loss": 0.0869, + "num_input_tokens_seen": 32076384, + "step": 24735 + }, + { + "epoch": 1.208829062125913, + "grad_norm": 0.0009279006044380367, + "learning_rate": 0.09545012219828314, + "loss": 0.0736, + "num_input_tokens_seen": 32082880, + "step": 24740 + }, + { + "epoch": 1.2090733637896074, + "grad_norm": 0.0014982393477112055, + "learning_rate": 0.09539525479088577, + "loss": 0.0635, + "num_input_tokens_seen": 32088928, + "step": 24745 + }, + { + "epoch": 1.2093176654533018, + "grad_norm": 0.0012659536441788077, + "learning_rate": 0.0953403958042264, + "loss": 0.1094, + "num_input_tokens_seen": 32095200, + "step": 24750 + }, + { + "epoch": 1.2095619671169962, + "grad_norm": 0.0007311746012419462, + "learning_rate": 0.09528554524676484, + "loss": 0.0714, + "num_input_tokens_seen": 32101856, + "step": 24755 + }, + { + "epoch": 1.2098062687806903, + "grad_norm": 0.0025652945041656494, + "learning_rate": 0.09523070312695978, + "loss": 0.1214, + "num_input_tokens_seen": 32108096, + "step": 24760 + }, + { + "epoch": 1.2100505704443847, + "grad_norm": 0.0016901622293516994, + "learning_rate": 0.09517586945326863, + "loss": 0.1031, + "num_input_tokens_seen": 32114304, + "step": 24765 + }, + { + "epoch": 1.210294872108079, + "grad_norm": 0.0022855361457914114, + "learning_rate": 0.0951210442341473, + "loss": 0.1047, + "num_input_tokens_seen": 32120576, + "step": 24770 + }, + { + "epoch": 1.2105391737717734, + "grad_norm": 0.0010300357826054096, + "learning_rate": 0.09506622747805066, + "loss": 0.0847, + "num_input_tokens_seen": 32126688, + "step": 24775 + }, + { + "epoch": 1.2107834754354676, + "grad_norm": 0.0011864943662658334, + "learning_rate": 0.09501141919343203, + "loss": 0.0949, + "num_input_tokens_seen": 32132864, + "step": 24780 + }, + { + "epoch": 1.211027777099162, + "grad_norm": 0.0012937304563820362, + "learning_rate": 0.09495661938874361, + "loss": 0.0846, + "num_input_tokens_seen": 32139136, + "step": 24785 + }, + { + "epoch": 1.2112720787628564, + "grad_norm": 0.0008229885715991259, + "learning_rate": 0.0949018280724362, + "loss": 0.0861, + "num_input_tokens_seen": 32145440, + "step": 24790 + }, + { + "epoch": 1.2115163804265507, + "grad_norm": 0.0014001497766003013, + "learning_rate": 0.09484704525295934, + "loss": 0.0863, + "num_input_tokens_seen": 32151936, + "step": 24795 + }, + { + "epoch": 1.2117606820902451, + "grad_norm": 0.0015727505087852478, + "learning_rate": 0.09479227093876112, + "loss": 0.1033, + "num_input_tokens_seen": 32158304, + "step": 24800 + }, + { + "epoch": 1.2117606820902451, + "eval_loss": 0.09121868759393692, + "eval_runtime": 402.6976, + "eval_samples_per_second": 90.353, + "eval_steps_per_second": 22.59, + "num_input_tokens_seen": 32158304, + "step": 24800 + }, + { + "epoch": 1.2120049837539393, + "grad_norm": 0.0015074346447363496, + "learning_rate": 0.0947375051382886, + "loss": 0.0908, + "num_input_tokens_seen": 32164832, + "step": 24805 + }, + { + "epoch": 1.2122492854176337, + "grad_norm": 0.0013095560716465116, + "learning_rate": 0.09468274785998718, + "loss": 0.074, + "num_input_tokens_seen": 32171360, + "step": 24810 + }, + { + "epoch": 1.212493587081328, + "grad_norm": 0.0010292710503563285, + "learning_rate": 0.09462799911230127, + "loss": 0.1027, + "num_input_tokens_seen": 32178016, + "step": 24815 + }, + { + "epoch": 1.2127378887450224, + "grad_norm": 0.0010619598906487226, + "learning_rate": 0.0945732589036737, + "loss": 0.0813, + "num_input_tokens_seen": 32184320, + "step": 24820 + }, + { + "epoch": 1.2129821904087166, + "grad_norm": 0.001236091135069728, + "learning_rate": 0.09451852724254614, + "loss": 0.1055, + "num_input_tokens_seen": 32190816, + "step": 24825 + }, + { + "epoch": 1.213226492072411, + "grad_norm": 0.001337636262178421, + "learning_rate": 0.09446380413735894, + "loss": 0.0789, + "num_input_tokens_seen": 32197440, + "step": 24830 + }, + { + "epoch": 1.2134707937361053, + "grad_norm": 0.0013081032084301114, + "learning_rate": 0.09440908959655099, + "loss": 0.0855, + "num_input_tokens_seen": 32204096, + "step": 24835 + }, + { + "epoch": 1.2137150953997997, + "grad_norm": 0.0011090370826423168, + "learning_rate": 0.09435438362856004, + "loss": 0.08, + "num_input_tokens_seen": 32210560, + "step": 24840 + }, + { + "epoch": 1.213959397063494, + "grad_norm": 0.0007479005143977702, + "learning_rate": 0.0942996862418225, + "loss": 0.0815, + "num_input_tokens_seen": 32217600, + "step": 24845 + }, + { + "epoch": 1.2142036987271883, + "grad_norm": 0.0013966754777356982, + "learning_rate": 0.09424499744477322, + "loss": 0.0625, + "num_input_tokens_seen": 32224160, + "step": 24850 + }, + { + "epoch": 1.2144480003908826, + "grad_norm": 0.001317319693043828, + "learning_rate": 0.09419031724584608, + "loss": 0.0727, + "num_input_tokens_seen": 32230560, + "step": 24855 + }, + { + "epoch": 1.214692302054577, + "grad_norm": 0.0010014886502176523, + "learning_rate": 0.09413564565347331, + "loss": 0.0753, + "num_input_tokens_seen": 32236736, + "step": 24860 + }, + { + "epoch": 1.2149366037182714, + "grad_norm": 0.0014766913373023272, + "learning_rate": 0.094080982676086, + "loss": 0.0871, + "num_input_tokens_seen": 32243200, + "step": 24865 + }, + { + "epoch": 1.2151809053819655, + "grad_norm": 0.0011078300885856152, + "learning_rate": 0.09402632832211395, + "loss": 0.0847, + "num_input_tokens_seen": 32249728, + "step": 24870 + }, + { + "epoch": 1.21542520704566, + "grad_norm": 0.001795479329302907, + "learning_rate": 0.09397168259998541, + "loss": 0.0953, + "num_input_tokens_seen": 32256416, + "step": 24875 + }, + { + "epoch": 1.2156695087093543, + "grad_norm": 0.0008391398587264121, + "learning_rate": 0.09391704551812759, + "loss": 0.0679, + "num_input_tokens_seen": 32262560, + "step": 24880 + }, + { + "epoch": 1.2159138103730487, + "grad_norm": 0.002071611350402236, + "learning_rate": 0.09386241708496605, + "loss": 0.0784, + "num_input_tokens_seen": 32268928, + "step": 24885 + }, + { + "epoch": 1.216158112036743, + "grad_norm": 0.0007873475551605225, + "learning_rate": 0.09380779730892527, + "loss": 0.0905, + "num_input_tokens_seen": 32275200, + "step": 24890 + }, + { + "epoch": 1.2164024137004372, + "grad_norm": 0.0010818374576047063, + "learning_rate": 0.09375318619842836, + "loss": 0.0845, + "num_input_tokens_seen": 32282080, + "step": 24895 + }, + { + "epoch": 1.2166467153641316, + "grad_norm": 0.0009420563001185656, + "learning_rate": 0.09369858376189696, + "loss": 0.0817, + "num_input_tokens_seen": 32289184, + "step": 24900 + }, + { + "epoch": 1.216891017027826, + "grad_norm": 0.0009436981636099517, + "learning_rate": 0.09364399000775143, + "loss": 0.1108, + "num_input_tokens_seen": 32295776, + "step": 24905 + }, + { + "epoch": 1.2171353186915204, + "grad_norm": 0.0016276647802442312, + "learning_rate": 0.09358940494441093, + "loss": 0.0741, + "num_input_tokens_seen": 32302240, + "step": 24910 + }, + { + "epoch": 1.2173796203552145, + "grad_norm": 0.0017380480421707034, + "learning_rate": 0.09353482858029301, + "loss": 0.0964, + "num_input_tokens_seen": 32308384, + "step": 24915 + }, + { + "epoch": 1.217623922018909, + "grad_norm": 0.001124605885706842, + "learning_rate": 0.09348026092381419, + "loss": 0.0891, + "num_input_tokens_seen": 32314912, + "step": 24920 + }, + { + "epoch": 1.2178682236826033, + "grad_norm": 0.0013288345653563738, + "learning_rate": 0.09342570198338931, + "loss": 0.0727, + "num_input_tokens_seen": 32322112, + "step": 24925 + }, + { + "epoch": 1.2181125253462977, + "grad_norm": 0.0008596524712629616, + "learning_rate": 0.0933711517674322, + "loss": 0.0811, + "num_input_tokens_seen": 32328704, + "step": 24930 + }, + { + "epoch": 1.218356827009992, + "grad_norm": 0.001164320157840848, + "learning_rate": 0.09331661028435513, + "loss": 0.0762, + "num_input_tokens_seen": 32335168, + "step": 24935 + }, + { + "epoch": 1.2186011286736862, + "grad_norm": 0.0010882950155064464, + "learning_rate": 0.09326207754256909, + "loss": 0.0596, + "num_input_tokens_seen": 32341856, + "step": 24940 + }, + { + "epoch": 1.2188454303373806, + "grad_norm": 0.0007956969202496111, + "learning_rate": 0.09320755355048366, + "loss": 0.0657, + "num_input_tokens_seen": 32348384, + "step": 24945 + }, + { + "epoch": 1.219089732001075, + "grad_norm": 0.0014512810157611966, + "learning_rate": 0.09315303831650722, + "loss": 0.0689, + "num_input_tokens_seen": 32355136, + "step": 24950 + }, + { + "epoch": 1.2193340336647693, + "grad_norm": 0.0006691979942843318, + "learning_rate": 0.09309853184904661, + "loss": 0.0606, + "num_input_tokens_seen": 32361568, + "step": 24955 + }, + { + "epoch": 1.2195783353284635, + "grad_norm": 0.0009472097735852003, + "learning_rate": 0.09304403415650753, + "loss": 0.0767, + "num_input_tokens_seen": 32368128, + "step": 24960 + }, + { + "epoch": 1.2198226369921579, + "grad_norm": 0.0011972495121881366, + "learning_rate": 0.09298954524729405, + "loss": 0.1077, + "num_input_tokens_seen": 32374336, + "step": 24965 + }, + { + "epoch": 1.2200669386558523, + "grad_norm": 0.0013085213722661138, + "learning_rate": 0.09293506512980916, + "loss": 0.0804, + "num_input_tokens_seen": 32381024, + "step": 24970 + }, + { + "epoch": 1.2203112403195466, + "grad_norm": 0.0013523672241717577, + "learning_rate": 0.0928805938124544, + "loss": 0.0814, + "num_input_tokens_seen": 32387552, + "step": 24975 + }, + { + "epoch": 1.220555541983241, + "grad_norm": 0.0013504025992006063, + "learning_rate": 0.09282613130362982, + "loss": 0.11, + "num_input_tokens_seen": 32394048, + "step": 24980 + }, + { + "epoch": 1.2207998436469352, + "grad_norm": 0.000633395160548389, + "learning_rate": 0.09277167761173427, + "loss": 0.0654, + "num_input_tokens_seen": 32400352, + "step": 24985 + }, + { + "epoch": 1.2210441453106295, + "grad_norm": 0.0010032793506979942, + "learning_rate": 0.0927172327451653, + "loss": 0.0794, + "num_input_tokens_seen": 32406944, + "step": 24990 + }, + { + "epoch": 1.221288446974324, + "grad_norm": 0.0012490125373005867, + "learning_rate": 0.09266279671231882, + "loss": 0.0975, + "num_input_tokens_seen": 32412672, + "step": 24995 + }, + { + "epoch": 1.2215327486380183, + "grad_norm": 0.0013059823540970683, + "learning_rate": 0.09260836952158967, + "loss": 0.0739, + "num_input_tokens_seen": 32419552, + "step": 25000 + }, + { + "epoch": 1.2215327486380183, + "eval_loss": 0.09187959879636765, + "eval_runtime": 402.6969, + "eval_samples_per_second": 90.353, + "eval_steps_per_second": 22.59, + "num_input_tokens_seen": 32419552, + "step": 25000 + }, + { + "epoch": 1.2217770503017125, + "grad_norm": 0.0007767357747070491, + "learning_rate": 0.09255395118137114, + "loss": 0.0923, + "num_input_tokens_seen": 32425728, + "step": 25005 + }, + { + "epoch": 1.2220213519654068, + "grad_norm": 0.000764586147852242, + "learning_rate": 0.09249954170005527, + "loss": 0.0919, + "num_input_tokens_seen": 32432096, + "step": 25010 + }, + { + "epoch": 1.2222656536291012, + "grad_norm": 0.0008679453167133033, + "learning_rate": 0.0924451410860327, + "loss": 0.1007, + "num_input_tokens_seen": 32438528, + "step": 25015 + }, + { + "epoch": 1.2225099552927956, + "grad_norm": 0.0008342025685124099, + "learning_rate": 0.09239074934769258, + "loss": 0.0879, + "num_input_tokens_seen": 32445632, + "step": 25020 + }, + { + "epoch": 1.22275425695649, + "grad_norm": 0.001683821901679039, + "learning_rate": 0.09233636649342288, + "loss": 0.0792, + "num_input_tokens_seen": 32451936, + "step": 25025 + }, + { + "epoch": 1.2229985586201841, + "grad_norm": 0.0008436593343503773, + "learning_rate": 0.09228199253161017, + "loss": 0.0889, + "num_input_tokens_seen": 32458176, + "step": 25030 + }, + { + "epoch": 1.2232428602838785, + "grad_norm": 0.0009199716150760651, + "learning_rate": 0.09222762747063949, + "loss": 0.0697, + "num_input_tokens_seen": 32464640, + "step": 25035 + }, + { + "epoch": 1.223487161947573, + "grad_norm": 0.0011252081021666527, + "learning_rate": 0.09217327131889473, + "loss": 0.0748, + "num_input_tokens_seen": 32471904, + "step": 25040 + }, + { + "epoch": 1.2237314636112673, + "grad_norm": 0.0012902957387268543, + "learning_rate": 0.09211892408475818, + "loss": 0.0925, + "num_input_tokens_seen": 32478240, + "step": 25045 + }, + { + "epoch": 1.2239757652749614, + "grad_norm": 0.0011525892186909914, + "learning_rate": 0.09206458577661089, + "loss": 0.0786, + "num_input_tokens_seen": 32484768, + "step": 25050 + }, + { + "epoch": 1.2242200669386558, + "grad_norm": 0.0013195305364206433, + "learning_rate": 0.09201025640283263, + "loss": 0.1016, + "num_input_tokens_seen": 32491264, + "step": 25055 + }, + { + "epoch": 1.2244643686023502, + "grad_norm": 0.0022841852623969316, + "learning_rate": 0.09195593597180148, + "loss": 0.1052, + "num_input_tokens_seen": 32497344, + "step": 25060 + }, + { + "epoch": 1.2247086702660446, + "grad_norm": 0.0014254512498155236, + "learning_rate": 0.09190162449189444, + "loss": 0.1052, + "num_input_tokens_seen": 32503680, + "step": 25065 + }, + { + "epoch": 1.224952971929739, + "grad_norm": 0.0010687249014154077, + "learning_rate": 0.09184732197148705, + "loss": 0.0737, + "num_input_tokens_seen": 32510016, + "step": 25070 + }, + { + "epoch": 1.2251972735934331, + "grad_norm": 0.0008551210048608482, + "learning_rate": 0.09179302841895343, + "loss": 0.0606, + "num_input_tokens_seen": 32516320, + "step": 25075 + }, + { + "epoch": 1.2254415752571275, + "grad_norm": 0.0009433721425011754, + "learning_rate": 0.09173874384266625, + "loss": 0.0815, + "num_input_tokens_seen": 32522400, + "step": 25080 + }, + { + "epoch": 1.2256858769208219, + "grad_norm": 0.0014181772712618113, + "learning_rate": 0.09168446825099695, + "loss": 0.0953, + "num_input_tokens_seen": 32528992, + "step": 25085 + }, + { + "epoch": 1.2259301785845163, + "grad_norm": 0.0018753400072455406, + "learning_rate": 0.09163020165231545, + "loss": 0.0917, + "num_input_tokens_seen": 32535296, + "step": 25090 + }, + { + "epoch": 1.2261744802482104, + "grad_norm": 0.0009280827362090349, + "learning_rate": 0.09157594405499044, + "loss": 0.0748, + "num_input_tokens_seen": 32541568, + "step": 25095 + }, + { + "epoch": 1.2264187819119048, + "grad_norm": 0.001698582898825407, + "learning_rate": 0.09152169546738899, + "loss": 0.0802, + "num_input_tokens_seen": 32548064, + "step": 25100 + }, + { + "epoch": 1.2266630835755992, + "grad_norm": 0.0009025080362334847, + "learning_rate": 0.09146745589787698, + "loss": 0.0648, + "num_input_tokens_seen": 32554592, + "step": 25105 + }, + { + "epoch": 1.2269073852392935, + "grad_norm": 0.000705883780028671, + "learning_rate": 0.09141322535481891, + "loss": 0.0727, + "num_input_tokens_seen": 32561440, + "step": 25110 + }, + { + "epoch": 1.227151686902988, + "grad_norm": 0.0010231444612145424, + "learning_rate": 0.0913590038465777, + "loss": 0.0707, + "num_input_tokens_seen": 32568064, + "step": 25115 + }, + { + "epoch": 1.227395988566682, + "grad_norm": 0.000992232351563871, + "learning_rate": 0.09130479138151505, + "loss": 0.0927, + "num_input_tokens_seen": 32574592, + "step": 25120 + }, + { + "epoch": 1.2276402902303765, + "grad_norm": 0.0008727959357202053, + "learning_rate": 0.09125058796799114, + "loss": 0.1, + "num_input_tokens_seen": 32580608, + "step": 25125 + }, + { + "epoch": 1.2278845918940708, + "grad_norm": 0.0018414370715618134, + "learning_rate": 0.09119639361436485, + "loss": 0.1039, + "num_input_tokens_seen": 32587200, + "step": 25130 + }, + { + "epoch": 1.2281288935577652, + "grad_norm": 0.0016356776468455791, + "learning_rate": 0.09114220832899368, + "loss": 0.1097, + "num_input_tokens_seen": 32593568, + "step": 25135 + }, + { + "epoch": 1.2283731952214594, + "grad_norm": 0.001007022918201983, + "learning_rate": 0.0910880321202336, + "loss": 0.0926, + "num_input_tokens_seen": 32600128, + "step": 25140 + }, + { + "epoch": 1.2286174968851538, + "grad_norm": 0.0008854287443682551, + "learning_rate": 0.09103386499643933, + "loss": 0.0734, + "num_input_tokens_seen": 32606336, + "step": 25145 + }, + { + "epoch": 1.2288617985488481, + "grad_norm": 0.0013021955965086818, + "learning_rate": 0.09097970696596407, + "loss": 0.0603, + "num_input_tokens_seen": 32613088, + "step": 25150 + }, + { + "epoch": 1.2291061002125425, + "grad_norm": 0.0008589564240537584, + "learning_rate": 0.09092555803715971, + "loss": 0.0959, + "num_input_tokens_seen": 32619520, + "step": 25155 + }, + { + "epoch": 1.229350401876237, + "grad_norm": 0.0028619577642530203, + "learning_rate": 0.0908714182183767, + "loss": 0.0859, + "num_input_tokens_seen": 32626016, + "step": 25160 + }, + { + "epoch": 1.229594703539931, + "grad_norm": 0.0014139015693217516, + "learning_rate": 0.090817287517964, + "loss": 0.0947, + "num_input_tokens_seen": 32632352, + "step": 25165 + }, + { + "epoch": 1.2298390052036254, + "grad_norm": 0.0010151856113225222, + "learning_rate": 0.09076316594426931, + "loss": 0.0706, + "num_input_tokens_seen": 32639552, + "step": 25170 + }, + { + "epoch": 1.2300833068673198, + "grad_norm": 0.0011669406667351723, + "learning_rate": 0.09070905350563888, + "loss": 0.0627, + "num_input_tokens_seen": 32646464, + "step": 25175 + }, + { + "epoch": 1.230327608531014, + "grad_norm": 0.0010708903428167105, + "learning_rate": 0.09065495021041745, + "loss": 0.0864, + "num_input_tokens_seen": 32653024, + "step": 25180 + }, + { + "epoch": 1.2305719101947084, + "grad_norm": 0.0011160590220242739, + "learning_rate": 0.09060085606694851, + "loss": 0.075, + "num_input_tokens_seen": 32659072, + "step": 25185 + }, + { + "epoch": 1.2308162118584027, + "grad_norm": 0.0012902953894808888, + "learning_rate": 0.09054677108357405, + "loss": 0.0813, + "num_input_tokens_seen": 32665152, + "step": 25190 + }, + { + "epoch": 1.231060513522097, + "grad_norm": 0.0006989336689002812, + "learning_rate": 0.09049269526863457, + "loss": 0.0786, + "num_input_tokens_seen": 32671872, + "step": 25195 + }, + { + "epoch": 1.2313048151857915, + "grad_norm": 0.0007782441098242998, + "learning_rate": 0.09043862863046935, + "loss": 0.058, + "num_input_tokens_seen": 32677888, + "step": 25200 + }, + { + "epoch": 1.2313048151857915, + "eval_loss": 0.09475168585777283, + "eval_runtime": 403.2138, + "eval_samples_per_second": 90.237, + "eval_steps_per_second": 22.561, + "num_input_tokens_seen": 32677888, + "step": 25200 + }, + { + "epoch": 1.2315491168494859, + "grad_norm": 0.001436862163245678, + "learning_rate": 0.09038457117741602, + "loss": 0.0737, + "num_input_tokens_seen": 32684576, + "step": 25205 + }, + { + "epoch": 1.23179341851318, + "grad_norm": 0.001481569604948163, + "learning_rate": 0.09033052291781099, + "loss": 0.0725, + "num_input_tokens_seen": 32690912, + "step": 25210 + }, + { + "epoch": 1.2320377201768744, + "grad_norm": 0.0015857625985518098, + "learning_rate": 0.09027648385998926, + "loss": 0.0801, + "num_input_tokens_seen": 32697376, + "step": 25215 + }, + { + "epoch": 1.2322820218405688, + "grad_norm": 0.0028160293586552143, + "learning_rate": 0.09022245401228417, + "loss": 0.0902, + "num_input_tokens_seen": 32703584, + "step": 25220 + }, + { + "epoch": 1.232526323504263, + "grad_norm": 0.0007734753889963031, + "learning_rate": 0.09016843338302792, + "loss": 0.0662, + "num_input_tokens_seen": 32709760, + "step": 25225 + }, + { + "epoch": 1.2327706251679573, + "grad_norm": 0.0014846650883555412, + "learning_rate": 0.09011442198055115, + "loss": 0.063, + "num_input_tokens_seen": 32715968, + "step": 25230 + }, + { + "epoch": 1.2330149268316517, + "grad_norm": 0.0015656540635973215, + "learning_rate": 0.09006041981318305, + "loss": 0.1107, + "num_input_tokens_seen": 32722368, + "step": 25235 + }, + { + "epoch": 1.233259228495346, + "grad_norm": 0.001151642412878573, + "learning_rate": 0.09000642688925149, + "loss": 0.1013, + "num_input_tokens_seen": 32729088, + "step": 25240 + }, + { + "epoch": 1.2335035301590405, + "grad_norm": 0.0015232277801260352, + "learning_rate": 0.0899524432170828, + "loss": 0.0897, + "num_input_tokens_seen": 32735328, + "step": 25245 + }, + { + "epoch": 1.2337478318227346, + "grad_norm": 0.0006389568443410099, + "learning_rate": 0.08989846880500196, + "loss": 0.1024, + "num_input_tokens_seen": 32742272, + "step": 25250 + }, + { + "epoch": 1.233992133486429, + "grad_norm": 0.0017363044898957014, + "learning_rate": 0.08984450366133256, + "loss": 0.0967, + "num_input_tokens_seen": 32748736, + "step": 25255 + }, + { + "epoch": 1.2342364351501234, + "grad_norm": 0.0017825959948822856, + "learning_rate": 0.08979054779439664, + "loss": 0.0924, + "num_input_tokens_seen": 32755360, + "step": 25260 + }, + { + "epoch": 1.2344807368138178, + "grad_norm": 0.00125342165119946, + "learning_rate": 0.08973660121251485, + "loss": 0.0833, + "num_input_tokens_seen": 32761824, + "step": 25265 + }, + { + "epoch": 1.234725038477512, + "grad_norm": 0.0013228252064436674, + "learning_rate": 0.08968266392400655, + "loss": 0.0831, + "num_input_tokens_seen": 32768384, + "step": 25270 + }, + { + "epoch": 1.2349693401412063, + "grad_norm": 0.0009553452837280929, + "learning_rate": 0.0896287359371894, + "loss": 0.068, + "num_input_tokens_seen": 32774464, + "step": 25275 + }, + { + "epoch": 1.2352136418049007, + "grad_norm": 0.001943731214851141, + "learning_rate": 0.08957481726037989, + "loss": 0.0787, + "num_input_tokens_seen": 32780704, + "step": 25280 + }, + { + "epoch": 1.235457943468595, + "grad_norm": 0.0011120858835056424, + "learning_rate": 0.08952090790189286, + "loss": 0.0913, + "num_input_tokens_seen": 32787392, + "step": 25285 + }, + { + "epoch": 1.2357022451322894, + "grad_norm": 0.0010460768826305866, + "learning_rate": 0.08946700787004187, + "loss": 0.1096, + "num_input_tokens_seen": 32793248, + "step": 25290 + }, + { + "epoch": 1.2359465467959836, + "grad_norm": 0.001382363261654973, + "learning_rate": 0.08941311717313899, + "loss": 0.0858, + "num_input_tokens_seen": 32799808, + "step": 25295 + }, + { + "epoch": 1.236190848459678, + "grad_norm": 0.0014813512098044157, + "learning_rate": 0.08935923581949483, + "loss": 0.091, + "num_input_tokens_seen": 32806464, + "step": 25300 + }, + { + "epoch": 1.2364351501233724, + "grad_norm": 0.0008698052261024714, + "learning_rate": 0.0893053638174185, + "loss": 0.0741, + "num_input_tokens_seen": 32812928, + "step": 25305 + }, + { + "epoch": 1.2366794517870667, + "grad_norm": 0.0011591344373300672, + "learning_rate": 0.0892515011752179, + "loss": 0.084, + "num_input_tokens_seen": 32819648, + "step": 25310 + }, + { + "epoch": 1.2369237534507609, + "grad_norm": 0.0014097238890826702, + "learning_rate": 0.08919764790119918, + "loss": 0.0914, + "num_input_tokens_seen": 32825984, + "step": 25315 + }, + { + "epoch": 1.2371680551144553, + "grad_norm": 0.0012756565120071173, + "learning_rate": 0.08914380400366727, + "loss": 0.0831, + "num_input_tokens_seen": 32832192, + "step": 25320 + }, + { + "epoch": 1.2374123567781496, + "grad_norm": 0.0007900788332335651, + "learning_rate": 0.08908996949092551, + "loss": 0.0785, + "num_input_tokens_seen": 32838784, + "step": 25325 + }, + { + "epoch": 1.237656658441844, + "grad_norm": 0.0014684309717267752, + "learning_rate": 0.08903614437127592, + "loss": 0.0938, + "num_input_tokens_seen": 32845152, + "step": 25330 + }, + { + "epoch": 1.2379009601055384, + "grad_norm": 0.0010414622956886888, + "learning_rate": 0.088982328653019, + "loss": 0.0952, + "num_input_tokens_seen": 32851712, + "step": 25335 + }, + { + "epoch": 1.2381452617692326, + "grad_norm": 0.001610913430340588, + "learning_rate": 0.0889285223444538, + "loss": 0.1257, + "num_input_tokens_seen": 32858080, + "step": 25340 + }, + { + "epoch": 1.238389563432927, + "grad_norm": 0.001506212865933776, + "learning_rate": 0.08887472545387787, + "loss": 0.0889, + "num_input_tokens_seen": 32864672, + "step": 25345 + }, + { + "epoch": 1.2386338650966213, + "grad_norm": 0.0013406771467998624, + "learning_rate": 0.08882093798958751, + "loss": 0.0687, + "num_input_tokens_seen": 32871296, + "step": 25350 + }, + { + "epoch": 1.2388781667603157, + "grad_norm": 0.001416159444488585, + "learning_rate": 0.08876715995987726, + "loss": 0.0887, + "num_input_tokens_seen": 32877792, + "step": 25355 + }, + { + "epoch": 1.2391224684240099, + "grad_norm": 0.0009312194888480008, + "learning_rate": 0.08871339137304052, + "loss": 0.0962, + "num_input_tokens_seen": 32884640, + "step": 25360 + }, + { + "epoch": 1.2393667700877042, + "grad_norm": 0.0020033109467476606, + "learning_rate": 0.0886596322373689, + "loss": 0.0858, + "num_input_tokens_seen": 32891648, + "step": 25365 + }, + { + "epoch": 1.2396110717513986, + "grad_norm": 0.0012098157312721014, + "learning_rate": 0.08860588256115293, + "loss": 0.0973, + "num_input_tokens_seen": 32898432, + "step": 25370 + }, + { + "epoch": 1.239855373415093, + "grad_norm": 0.0010559025686234236, + "learning_rate": 0.0885521423526814, + "loss": 0.0855, + "num_input_tokens_seen": 32904512, + "step": 25375 + }, + { + "epoch": 1.2400996750787874, + "grad_norm": 0.0019284236477687955, + "learning_rate": 0.08849841162024165, + "loss": 0.1045, + "num_input_tokens_seen": 32910656, + "step": 25380 + }, + { + "epoch": 1.2403439767424815, + "grad_norm": 0.0015068593202158809, + "learning_rate": 0.08844469037211973, + "loss": 0.1262, + "num_input_tokens_seen": 32917216, + "step": 25385 + }, + { + "epoch": 1.240588278406176, + "grad_norm": 0.0006021906156092882, + "learning_rate": 0.08839097861660014, + "loss": 0.1043, + "num_input_tokens_seen": 32923808, + "step": 25390 + }, + { + "epoch": 1.2408325800698703, + "grad_norm": 0.0005265615764074028, + "learning_rate": 0.08833727636196585, + "loss": 0.0722, + "num_input_tokens_seen": 32930240, + "step": 25395 + }, + { + "epoch": 1.2410768817335647, + "grad_norm": 0.0013744068564847112, + "learning_rate": 0.08828358361649848, + "loss": 0.0679, + "num_input_tokens_seen": 32936608, + "step": 25400 + }, + { + "epoch": 1.2410768817335647, + "eval_loss": 0.09156373143196106, + "eval_runtime": 402.5231, + "eval_samples_per_second": 90.392, + "eval_steps_per_second": 22.6, + "num_input_tokens_seen": 32936608, + "step": 25400 + }, + { + "epoch": 1.2413211833972588, + "grad_norm": 0.0011855678167194128, + "learning_rate": 0.08822990038847807, + "loss": 0.0794, + "num_input_tokens_seen": 32942912, + "step": 25405 + }, + { + "epoch": 1.2415654850609532, + "grad_norm": 0.0011288790265098214, + "learning_rate": 0.08817622668618325, + "loss": 0.0695, + "num_input_tokens_seen": 32948800, + "step": 25410 + }, + { + "epoch": 1.2418097867246476, + "grad_norm": 0.0021093408577144146, + "learning_rate": 0.08812256251789125, + "loss": 0.1074, + "num_input_tokens_seen": 32955456, + "step": 25415 + }, + { + "epoch": 1.242054088388342, + "grad_norm": 0.0014707270311191678, + "learning_rate": 0.08806890789187766, + "loss": 0.0956, + "num_input_tokens_seen": 32961664, + "step": 25420 + }, + { + "epoch": 1.2422983900520363, + "grad_norm": 0.0011298621539026499, + "learning_rate": 0.08801526281641672, + "loss": 0.0831, + "num_input_tokens_seen": 32967872, + "step": 25425 + }, + { + "epoch": 1.2425426917157305, + "grad_norm": 0.0011873102048411965, + "learning_rate": 0.0879616272997813, + "loss": 0.0931, + "num_input_tokens_seen": 32973696, + "step": 25430 + }, + { + "epoch": 1.2427869933794249, + "grad_norm": 0.0010329593205824494, + "learning_rate": 0.08790800135024247, + "loss": 0.0711, + "num_input_tokens_seen": 32979904, + "step": 25435 + }, + { + "epoch": 1.2430312950431193, + "grad_norm": 0.0017863124376162887, + "learning_rate": 0.08785438497607023, + "loss": 0.1057, + "num_input_tokens_seen": 32986016, + "step": 25440 + }, + { + "epoch": 1.2432755967068136, + "grad_norm": 0.0008694807766005397, + "learning_rate": 0.08780077818553277, + "loss": 0.0737, + "num_input_tokens_seen": 32992160, + "step": 25445 + }, + { + "epoch": 1.2435198983705078, + "grad_norm": 0.0013092070585116744, + "learning_rate": 0.0877471809868969, + "loss": 0.1076, + "num_input_tokens_seen": 32998272, + "step": 25450 + }, + { + "epoch": 1.2437642000342022, + "grad_norm": 0.0013548372080549598, + "learning_rate": 0.08769359338842811, + "loss": 0.0885, + "num_input_tokens_seen": 33005344, + "step": 25455 + }, + { + "epoch": 1.2440085016978966, + "grad_norm": 0.0016625917050987482, + "learning_rate": 0.08764001539839016, + "loss": 0.1123, + "num_input_tokens_seen": 33011680, + "step": 25460 + }, + { + "epoch": 1.244252803361591, + "grad_norm": 0.0007802696782164276, + "learning_rate": 0.08758644702504548, + "loss": 0.0687, + "num_input_tokens_seen": 33018208, + "step": 25465 + }, + { + "epoch": 1.2444971050252853, + "grad_norm": 0.0005167615599930286, + "learning_rate": 0.0875328882766551, + "loss": 0.0728, + "num_input_tokens_seen": 33024768, + "step": 25470 + }, + { + "epoch": 1.2447414066889795, + "grad_norm": 0.0015481181908398867, + "learning_rate": 0.08747933916147828, + "loss": 0.0673, + "num_input_tokens_seen": 33031392, + "step": 25475 + }, + { + "epoch": 1.2449857083526739, + "grad_norm": 0.0006230099243111908, + "learning_rate": 0.0874257996877731, + "loss": 0.0701, + "num_input_tokens_seen": 33038304, + "step": 25480 + }, + { + "epoch": 1.2452300100163682, + "grad_norm": 0.0008590742945671082, + "learning_rate": 0.08737226986379593, + "loss": 0.093, + "num_input_tokens_seen": 33044736, + "step": 25485 + }, + { + "epoch": 1.2454743116800626, + "grad_norm": 0.00047651882050558925, + "learning_rate": 0.08731874969780173, + "loss": 0.0817, + "num_input_tokens_seen": 33051168, + "step": 25490 + }, + { + "epoch": 1.2457186133437568, + "grad_norm": 0.0017387480475008488, + "learning_rate": 0.08726523919804412, + "loss": 0.0946, + "num_input_tokens_seen": 33057792, + "step": 25495 + }, + { + "epoch": 1.2459629150074512, + "grad_norm": 0.0008188495994545519, + "learning_rate": 0.08721173837277492, + "loss": 0.0717, + "num_input_tokens_seen": 33064576, + "step": 25500 + }, + { + "epoch": 1.2462072166711455, + "grad_norm": 0.0019482828211039305, + "learning_rate": 0.08715824723024479, + "loss": 0.1206, + "num_input_tokens_seen": 33071360, + "step": 25505 + }, + { + "epoch": 1.24645151833484, + "grad_norm": 0.0005999520653858781, + "learning_rate": 0.08710476577870258, + "loss": 0.0701, + "num_input_tokens_seen": 33078080, + "step": 25510 + }, + { + "epoch": 1.2466958199985343, + "grad_norm": 0.00070715177571401, + "learning_rate": 0.08705129402639587, + "loss": 0.0922, + "num_input_tokens_seen": 33084448, + "step": 25515 + }, + { + "epoch": 1.2469401216622285, + "grad_norm": 0.0013087376719340682, + "learning_rate": 0.08699783198157078, + "loss": 0.0795, + "num_input_tokens_seen": 33090624, + "step": 25520 + }, + { + "epoch": 1.2471844233259228, + "grad_norm": 0.0009672612650319934, + "learning_rate": 0.08694437965247163, + "loss": 0.0702, + "num_input_tokens_seen": 33097440, + "step": 25525 + }, + { + "epoch": 1.2474287249896172, + "grad_norm": 0.0017823674716055393, + "learning_rate": 0.08689093704734165, + "loss": 0.1043, + "num_input_tokens_seen": 33104096, + "step": 25530 + }, + { + "epoch": 1.2476730266533116, + "grad_norm": 0.0004160506359767169, + "learning_rate": 0.08683750417442222, + "loss": 0.0727, + "num_input_tokens_seen": 33110848, + "step": 25535 + }, + { + "epoch": 1.2479173283170057, + "grad_norm": 0.0013404875062406063, + "learning_rate": 0.08678408104195334, + "loss": 0.0833, + "num_input_tokens_seen": 33117408, + "step": 25540 + }, + { + "epoch": 1.2481616299807001, + "grad_norm": 0.0013267146423459053, + "learning_rate": 0.08673066765817365, + "loss": 0.0872, + "num_input_tokens_seen": 33123904, + "step": 25545 + }, + { + "epoch": 1.2484059316443945, + "grad_norm": 0.0017367592081427574, + "learning_rate": 0.08667726403132005, + "loss": 0.062, + "num_input_tokens_seen": 33130208, + "step": 25550 + }, + { + "epoch": 1.2486502333080889, + "grad_norm": 0.0032236981205642223, + "learning_rate": 0.0866238701696281, + "loss": 0.0906, + "num_input_tokens_seen": 33136672, + "step": 25555 + }, + { + "epoch": 1.2488945349717833, + "grad_norm": 0.0021490200888365507, + "learning_rate": 0.08657048608133185, + "loss": 0.084, + "num_input_tokens_seen": 33142944, + "step": 25560 + }, + { + "epoch": 1.2491388366354774, + "grad_norm": 0.0011032845359295607, + "learning_rate": 0.08651711177466369, + "loss": 0.063, + "num_input_tokens_seen": 33149728, + "step": 25565 + }, + { + "epoch": 1.2493831382991718, + "grad_norm": 0.0014601285802200437, + "learning_rate": 0.08646374725785466, + "loss": 0.0997, + "num_input_tokens_seen": 33156576, + "step": 25570 + }, + { + "epoch": 1.2496274399628662, + "grad_norm": 0.001423213048838079, + "learning_rate": 0.08641039253913434, + "loss": 0.1141, + "num_input_tokens_seen": 33163328, + "step": 25575 + }, + { + "epoch": 1.2498717416265606, + "grad_norm": 0.0011598268756642938, + "learning_rate": 0.08635704762673052, + "loss": 0.0944, + "num_input_tokens_seen": 33169632, + "step": 25580 + }, + { + "epoch": 1.2501160432902547, + "grad_norm": 0.0011487844167277217, + "learning_rate": 0.08630371252886981, + "loss": 0.0862, + "num_input_tokens_seen": 33176224, + "step": 25585 + }, + { + "epoch": 1.250360344953949, + "grad_norm": 0.0013225860893726349, + "learning_rate": 0.08625038725377704, + "loss": 0.102, + "num_input_tokens_seen": 33182304, + "step": 25590 + }, + { + "epoch": 1.2506046466176435, + "grad_norm": 0.0007141039823181927, + "learning_rate": 0.08619707180967566, + "loss": 0.0778, + "num_input_tokens_seen": 33188896, + "step": 25595 + }, + { + "epoch": 1.2508489482813379, + "grad_norm": 0.0014052769402042031, + "learning_rate": 0.08614376620478768, + "loss": 0.1096, + "num_input_tokens_seen": 33195264, + "step": 25600 + }, + { + "epoch": 1.2508489482813379, + "eval_loss": 0.09638015925884247, + "eval_runtime": 402.5507, + "eval_samples_per_second": 90.386, + "eval_steps_per_second": 22.598, + "num_input_tokens_seen": 33195264, + "step": 25600 + }, + { + "epoch": 1.2510932499450322, + "grad_norm": 0.0009300444507971406, + "learning_rate": 0.08609047044733344, + "loss": 0.0964, + "num_input_tokens_seen": 33201920, + "step": 25605 + }, + { + "epoch": 1.2513375516087264, + "grad_norm": 0.0011659222654998302, + "learning_rate": 0.08603718454553168, + "loss": 0.0908, + "num_input_tokens_seen": 33208192, + "step": 25610 + }, + { + "epoch": 1.2515818532724208, + "grad_norm": 0.001170493895187974, + "learning_rate": 0.08598390850759997, + "loss": 0.0975, + "num_input_tokens_seen": 33214624, + "step": 25615 + }, + { + "epoch": 1.2518261549361152, + "grad_norm": 0.0013358592987060547, + "learning_rate": 0.08593064234175397, + "loss": 0.0854, + "num_input_tokens_seen": 33220736, + "step": 25620 + }, + { + "epoch": 1.2520704565998093, + "grad_norm": 0.0012277048081159592, + "learning_rate": 0.08587738605620815, + "loss": 0.0823, + "num_input_tokens_seen": 33227584, + "step": 25625 + }, + { + "epoch": 1.2523147582635037, + "grad_norm": 0.0015130293322727084, + "learning_rate": 0.08582413965917512, + "loss": 0.0897, + "num_input_tokens_seen": 33234400, + "step": 25630 + }, + { + "epoch": 1.252559059927198, + "grad_norm": 0.0011364771053195, + "learning_rate": 0.08577090315886628, + "loss": 0.0806, + "num_input_tokens_seen": 33241152, + "step": 25635 + }, + { + "epoch": 1.2528033615908925, + "grad_norm": 0.0008847310091368854, + "learning_rate": 0.08571767656349136, + "loss": 0.0874, + "num_input_tokens_seen": 33247520, + "step": 25640 + }, + { + "epoch": 1.2530476632545868, + "grad_norm": 0.000741969037335366, + "learning_rate": 0.08566445988125847, + "loss": 0.0782, + "num_input_tokens_seen": 33254304, + "step": 25645 + }, + { + "epoch": 1.2532919649182812, + "grad_norm": 0.0008664630586281419, + "learning_rate": 0.08561125312037436, + "loss": 0.0606, + "num_input_tokens_seen": 33260928, + "step": 25650 + }, + { + "epoch": 1.2535362665819754, + "grad_norm": 0.0014305372023954988, + "learning_rate": 0.08555805628904424, + "loss": 0.1103, + "num_input_tokens_seen": 33267264, + "step": 25655 + }, + { + "epoch": 1.2537805682456697, + "grad_norm": 0.000922060979064554, + "learning_rate": 0.08550486939547161, + "loss": 0.0882, + "num_input_tokens_seen": 33273536, + "step": 25660 + }, + { + "epoch": 1.2540248699093641, + "grad_norm": 0.0010172042530030012, + "learning_rate": 0.08545169244785869, + "loss": 0.0904, + "num_input_tokens_seen": 33280256, + "step": 25665 + }, + { + "epoch": 1.2542691715730583, + "grad_norm": 0.0010000189067795873, + "learning_rate": 0.08539852545440589, + "loss": 0.0824, + "num_input_tokens_seen": 33286944, + "step": 25670 + }, + { + "epoch": 1.2545134732367527, + "grad_norm": 0.0009306779247708619, + "learning_rate": 0.08534536842331235, + "loss": 0.0933, + "num_input_tokens_seen": 33293312, + "step": 25675 + }, + { + "epoch": 1.254757774900447, + "grad_norm": 0.0010808005463331938, + "learning_rate": 0.08529222136277545, + "loss": 0.057, + "num_input_tokens_seen": 33299968, + "step": 25680 + }, + { + "epoch": 1.2550020765641414, + "grad_norm": 0.0014186814660206437, + "learning_rate": 0.08523908428099125, + "loss": 0.0901, + "num_input_tokens_seen": 33306272, + "step": 25685 + }, + { + "epoch": 1.2552463782278358, + "grad_norm": 0.001213140320032835, + "learning_rate": 0.08518595718615402, + "loss": 0.0942, + "num_input_tokens_seen": 33313120, + "step": 25690 + }, + { + "epoch": 1.2554906798915302, + "grad_norm": 0.0009904083563014865, + "learning_rate": 0.08513284008645675, + "loss": 0.0923, + "num_input_tokens_seen": 33319040, + "step": 25695 + }, + { + "epoch": 1.2557349815552243, + "grad_norm": 0.001651423517614603, + "learning_rate": 0.08507973299009065, + "loss": 0.0953, + "num_input_tokens_seen": 33325376, + "step": 25700 + }, + { + "epoch": 1.2559792832189187, + "grad_norm": 0.0012850068742409348, + "learning_rate": 0.08502663590524563, + "loss": 0.088, + "num_input_tokens_seen": 33332352, + "step": 25705 + }, + { + "epoch": 1.256223584882613, + "grad_norm": 0.0011182768503203988, + "learning_rate": 0.08497354884010981, + "loss": 0.1145, + "num_input_tokens_seen": 33339104, + "step": 25710 + }, + { + "epoch": 1.2564678865463073, + "grad_norm": 0.0011601962614804506, + "learning_rate": 0.0849204718028699, + "loss": 0.0622, + "num_input_tokens_seen": 33346176, + "step": 25715 + }, + { + "epoch": 1.2567121882100016, + "grad_norm": 0.000826326955575496, + "learning_rate": 0.08486740480171118, + "loss": 0.0952, + "num_input_tokens_seen": 33352448, + "step": 25720 + }, + { + "epoch": 1.256956489873696, + "grad_norm": 0.0013120108051225543, + "learning_rate": 0.08481434784481706, + "loss": 0.0841, + "num_input_tokens_seen": 33359040, + "step": 25725 + }, + { + "epoch": 1.2572007915373904, + "grad_norm": 0.0011952818604186177, + "learning_rate": 0.08476130094036968, + "loss": 0.0857, + "num_input_tokens_seen": 33365408, + "step": 25730 + }, + { + "epoch": 1.2574450932010848, + "grad_norm": 0.0009078708244487643, + "learning_rate": 0.08470826409654961, + "loss": 0.069, + "num_input_tokens_seen": 33372000, + "step": 25735 + }, + { + "epoch": 1.2576893948647792, + "grad_norm": 0.001870551728643477, + "learning_rate": 0.08465523732153564, + "loss": 0.089, + "num_input_tokens_seen": 33378560, + "step": 25740 + }, + { + "epoch": 1.2579336965284733, + "grad_norm": 0.0014545354060828686, + "learning_rate": 0.08460222062350532, + "loss": 0.0755, + "num_input_tokens_seen": 33384992, + "step": 25745 + }, + { + "epoch": 1.2581779981921677, + "grad_norm": 0.0008357822662219405, + "learning_rate": 0.08454921401063442, + "loss": 0.0829, + "num_input_tokens_seen": 33391424, + "step": 25750 + }, + { + "epoch": 1.258422299855862, + "grad_norm": 0.0021431976929306984, + "learning_rate": 0.08449621749109716, + "loss": 0.0762, + "num_input_tokens_seen": 33397888, + "step": 25755 + }, + { + "epoch": 1.2586666015195562, + "grad_norm": 0.0009551410912536085, + "learning_rate": 0.08444323107306641, + "loss": 0.0851, + "num_input_tokens_seen": 33404128, + "step": 25760 + }, + { + "epoch": 1.2589109031832506, + "grad_norm": 0.0009083696641027927, + "learning_rate": 0.0843902547647132, + "loss": 0.108, + "num_input_tokens_seen": 33410464, + "step": 25765 + }, + { + "epoch": 1.259155204846945, + "grad_norm": 0.0021048772614449263, + "learning_rate": 0.0843372885742072, + "loss": 0.112, + "num_input_tokens_seen": 33416864, + "step": 25770 + }, + { + "epoch": 1.2593995065106394, + "grad_norm": 0.0010716529795899987, + "learning_rate": 0.08428433250971652, + "loss": 0.0898, + "num_input_tokens_seen": 33423328, + "step": 25775 + }, + { + "epoch": 1.2596438081743337, + "grad_norm": 0.0011246780632063746, + "learning_rate": 0.08423138657940757, + "loss": 0.0935, + "num_input_tokens_seen": 33429600, + "step": 25780 + }, + { + "epoch": 1.2598881098380281, + "grad_norm": 0.0009619712363928556, + "learning_rate": 0.08417845079144536, + "loss": 0.1181, + "num_input_tokens_seen": 33435552, + "step": 25785 + }, + { + "epoch": 1.2601324115017223, + "grad_norm": 0.001313935499638319, + "learning_rate": 0.08412552515399314, + "loss": 0.0866, + "num_input_tokens_seen": 33441696, + "step": 25790 + }, + { + "epoch": 1.2603767131654167, + "grad_norm": 0.0019261217676103115, + "learning_rate": 0.08407260967521278, + "loss": 0.0779, + "num_input_tokens_seen": 33448256, + "step": 25795 + }, + { + "epoch": 1.260621014829111, + "grad_norm": 0.0013964304234832525, + "learning_rate": 0.08401970436326454, + "loss": 0.0767, + "num_input_tokens_seen": 33454720, + "step": 25800 + }, + { + "epoch": 1.260621014829111, + "eval_loss": 0.0945761427283287, + "eval_runtime": 402.156, + "eval_samples_per_second": 90.475, + "eval_steps_per_second": 22.621, + "num_input_tokens_seen": 33454720, + "step": 25800 + }, + { + "epoch": 1.2608653164928052, + "grad_norm": 0.0014390724245458841, + "learning_rate": 0.08396680922630702, + "loss": 0.0948, + "num_input_tokens_seen": 33460992, + "step": 25805 + }, + { + "epoch": 1.2611096181564996, + "grad_norm": 0.0010058843763545156, + "learning_rate": 0.08391392427249732, + "loss": 0.0743, + "num_input_tokens_seen": 33467040, + "step": 25810 + }, + { + "epoch": 1.261353919820194, + "grad_norm": 0.0011219828156754375, + "learning_rate": 0.08386104950999107, + "loss": 0.0879, + "num_input_tokens_seen": 33473248, + "step": 25815 + }, + { + "epoch": 1.2615982214838883, + "grad_norm": 0.001160880783572793, + "learning_rate": 0.0838081849469421, + "loss": 0.0894, + "num_input_tokens_seen": 33479776, + "step": 25820 + }, + { + "epoch": 1.2618425231475827, + "grad_norm": 0.001041596056893468, + "learning_rate": 0.08375533059150281, + "loss": 0.0955, + "num_input_tokens_seen": 33486144, + "step": 25825 + }, + { + "epoch": 1.262086824811277, + "grad_norm": 0.0007353362161666155, + "learning_rate": 0.08370248645182406, + "loss": 0.0662, + "num_input_tokens_seen": 33493600, + "step": 25830 + }, + { + "epoch": 1.2623311264749713, + "grad_norm": 0.000903365551494062, + "learning_rate": 0.083649652536055, + "loss": 0.0514, + "num_input_tokens_seen": 33499872, + "step": 25835 + }, + { + "epoch": 1.2625754281386656, + "grad_norm": 0.0008618325809948146, + "learning_rate": 0.08359682885234339, + "loss": 0.0896, + "num_input_tokens_seen": 33506848, + "step": 25840 + }, + { + "epoch": 1.26281972980236, + "grad_norm": 0.0013618952361866832, + "learning_rate": 0.08354401540883516, + "loss": 0.0826, + "num_input_tokens_seen": 33513440, + "step": 25845 + }, + { + "epoch": 1.2630640314660542, + "grad_norm": 0.0014080960536375642, + "learning_rate": 0.0834912122136749, + "loss": 0.0999, + "num_input_tokens_seen": 33519872, + "step": 25850 + }, + { + "epoch": 1.2633083331297486, + "grad_norm": 0.0008109242771752179, + "learning_rate": 0.0834384192750056, + "loss": 0.1001, + "num_input_tokens_seen": 33526496, + "step": 25855 + }, + { + "epoch": 1.263552634793443, + "grad_norm": 0.0014026208082213998, + "learning_rate": 0.08338563660096844, + "loss": 0.0987, + "num_input_tokens_seen": 33532928, + "step": 25860 + }, + { + "epoch": 1.2637969364571373, + "grad_norm": 0.001026695012114942, + "learning_rate": 0.08333286419970329, + "loss": 0.0947, + "num_input_tokens_seen": 33539968, + "step": 25865 + }, + { + "epoch": 1.2640412381208317, + "grad_norm": 0.0012038060231134295, + "learning_rate": 0.08328010207934824, + "loss": 0.103, + "num_input_tokens_seen": 33546048, + "step": 25870 + }, + { + "epoch": 1.264285539784526, + "grad_norm": 0.0014178537530824542, + "learning_rate": 0.08322735024803989, + "loss": 0.0722, + "num_input_tokens_seen": 33552800, + "step": 25875 + }, + { + "epoch": 1.2645298414482202, + "grad_norm": 0.0011038953671231866, + "learning_rate": 0.08317460871391331, + "loss": 0.0838, + "num_input_tokens_seen": 33559712, + "step": 25880 + }, + { + "epoch": 1.2647741431119146, + "grad_norm": 0.0016846568323671818, + "learning_rate": 0.08312187748510179, + "loss": 0.1087, + "num_input_tokens_seen": 33565888, + "step": 25885 + }, + { + "epoch": 1.265018444775609, + "grad_norm": 0.001323905074968934, + "learning_rate": 0.08306915656973726, + "loss": 0.0961, + "num_input_tokens_seen": 33572256, + "step": 25890 + }, + { + "epoch": 1.2652627464393031, + "grad_norm": 0.0011115256929770112, + "learning_rate": 0.08301644597594988, + "loss": 0.0821, + "num_input_tokens_seen": 33578656, + "step": 25895 + }, + { + "epoch": 1.2655070481029975, + "grad_norm": 0.0014240107266232371, + "learning_rate": 0.08296374571186826, + "loss": 0.091, + "num_input_tokens_seen": 33584640, + "step": 25900 + }, + { + "epoch": 1.265751349766692, + "grad_norm": 0.001274219946935773, + "learning_rate": 0.08291105578561955, + "loss": 0.1045, + "num_input_tokens_seen": 33591264, + "step": 25905 + }, + { + "epoch": 1.2659956514303863, + "grad_norm": 0.0018116901628673077, + "learning_rate": 0.08285837620532904, + "loss": 0.0908, + "num_input_tokens_seen": 33597376, + "step": 25910 + }, + { + "epoch": 1.2662399530940807, + "grad_norm": 0.0013798102736473083, + "learning_rate": 0.0828057069791207, + "loss": 0.0692, + "num_input_tokens_seen": 33604000, + "step": 25915 + }, + { + "epoch": 1.2664842547577748, + "grad_norm": 0.0007336615235544741, + "learning_rate": 0.0827530481151168, + "loss": 0.0869, + "num_input_tokens_seen": 33610912, + "step": 25920 + }, + { + "epoch": 1.2667285564214692, + "grad_norm": 0.000923492421861738, + "learning_rate": 0.08270039962143792, + "loss": 0.0799, + "num_input_tokens_seen": 33617792, + "step": 25925 + }, + { + "epoch": 1.2669728580851636, + "grad_norm": 0.0005231878021731973, + "learning_rate": 0.08264776150620314, + "loss": 0.0642, + "num_input_tokens_seen": 33624448, + "step": 25930 + }, + { + "epoch": 1.267217159748858, + "grad_norm": 0.0013380820164456964, + "learning_rate": 0.08259513377753, + "loss": 0.1121, + "num_input_tokens_seen": 33630976, + "step": 25935 + }, + { + "epoch": 1.2674614614125521, + "grad_norm": 0.0010028143879026175, + "learning_rate": 0.08254251644353423, + "loss": 0.1008, + "num_input_tokens_seen": 33637280, + "step": 25940 + }, + { + "epoch": 1.2677057630762465, + "grad_norm": 0.0009190183482132852, + "learning_rate": 0.08248990951233022, + "loss": 0.0827, + "num_input_tokens_seen": 33644160, + "step": 25945 + }, + { + "epoch": 1.2679500647399409, + "grad_norm": 0.0009093956323340535, + "learning_rate": 0.08243731299203048, + "loss": 0.0804, + "num_input_tokens_seen": 33650528, + "step": 25950 + }, + { + "epoch": 1.2681943664036353, + "grad_norm": 0.0019930400885641575, + "learning_rate": 0.08238472689074612, + "loss": 0.0823, + "num_input_tokens_seen": 33657024, + "step": 25955 + }, + { + "epoch": 1.2684386680673296, + "grad_norm": 0.0012036616681143641, + "learning_rate": 0.08233215121658666, + "loss": 0.0973, + "num_input_tokens_seen": 33663296, + "step": 25960 + }, + { + "epoch": 1.2686829697310238, + "grad_norm": 0.0011712113628163934, + "learning_rate": 0.08227958597765982, + "loss": 0.0902, + "num_input_tokens_seen": 33669568, + "step": 25965 + }, + { + "epoch": 1.2689272713947182, + "grad_norm": 0.0011071496410295367, + "learning_rate": 0.08222703118207181, + "loss": 0.0876, + "num_input_tokens_seen": 33676000, + "step": 25970 + }, + { + "epoch": 1.2691715730584125, + "grad_norm": 0.0022062119096517563, + "learning_rate": 0.08217448683792734, + "loss": 0.0802, + "num_input_tokens_seen": 33682432, + "step": 25975 + }, + { + "epoch": 1.269415874722107, + "grad_norm": 0.0009557584417052567, + "learning_rate": 0.08212195295332926, + "loss": 0.062, + "num_input_tokens_seen": 33688928, + "step": 25980 + }, + { + "epoch": 1.269660176385801, + "grad_norm": 0.0019330477807670832, + "learning_rate": 0.08206942953637915, + "loss": 0.0889, + "num_input_tokens_seen": 33695456, + "step": 25985 + }, + { + "epoch": 1.2699044780494955, + "grad_norm": 0.001016924623399973, + "learning_rate": 0.08201691659517658, + "loss": 0.0677, + "num_input_tokens_seen": 33702208, + "step": 25990 + }, + { + "epoch": 1.2701487797131898, + "grad_norm": 0.0010259926784783602, + "learning_rate": 0.08196441413781981, + "loss": 0.1001, + "num_input_tokens_seen": 33708480, + "step": 25995 + }, + { + "epoch": 1.2703930813768842, + "grad_norm": 0.0012223697267472744, + "learning_rate": 0.08191192217240544, + "loss": 0.0938, + "num_input_tokens_seen": 33714496, + "step": 26000 + }, + { + "epoch": 1.2703930813768842, + "eval_loss": 0.0894274115562439, + "eval_runtime": 402.2967, + "eval_samples_per_second": 90.443, + "eval_steps_per_second": 22.613, + "num_input_tokens_seen": 33714496, + "step": 26000 + }, + { + "epoch": 1.2706373830405786, + "grad_norm": 0.0008279799367301166, + "learning_rate": 0.08185944070702823, + "loss": 0.0868, + "num_input_tokens_seen": 33720992, + "step": 26005 + }, + { + "epoch": 1.2708816847042728, + "grad_norm": 0.0020063985139131546, + "learning_rate": 0.08180696974978159, + "loss": 0.0886, + "num_input_tokens_seen": 33727680, + "step": 26010 + }, + { + "epoch": 1.2711259863679671, + "grad_norm": 0.0010507418774068356, + "learning_rate": 0.08175450930875724, + "loss": 0.0946, + "num_input_tokens_seen": 33733984, + "step": 26015 + }, + { + "epoch": 1.2713702880316615, + "grad_norm": 0.0011688568629324436, + "learning_rate": 0.08170205939204513, + "loss": 0.1047, + "num_input_tokens_seen": 33740384, + "step": 26020 + }, + { + "epoch": 1.271614589695356, + "grad_norm": 0.0010254436638206244, + "learning_rate": 0.08164962000773379, + "loss": 0.0883, + "num_input_tokens_seen": 33746848, + "step": 26025 + }, + { + "epoch": 1.27185889135905, + "grad_norm": 0.0005665483186021447, + "learning_rate": 0.08159719116390995, + "loss": 0.0857, + "num_input_tokens_seen": 33753024, + "step": 26030 + }, + { + "epoch": 1.2721031930227444, + "grad_norm": 0.0013368806103244424, + "learning_rate": 0.08154477286865887, + "loss": 0.0717, + "num_input_tokens_seen": 33759360, + "step": 26035 + }, + { + "epoch": 1.2723474946864388, + "grad_norm": 0.0009778544772416353, + "learning_rate": 0.08149236513006404, + "loss": 0.1009, + "num_input_tokens_seen": 33765632, + "step": 26040 + }, + { + "epoch": 1.2725917963501332, + "grad_norm": 0.001533033442683518, + "learning_rate": 0.08143996795620746, + "loss": 0.0918, + "num_input_tokens_seen": 33771872, + "step": 26045 + }, + { + "epoch": 1.2728360980138276, + "grad_norm": 0.0010330852819606662, + "learning_rate": 0.08138758135516938, + "loss": 0.0831, + "num_input_tokens_seen": 33778752, + "step": 26050 + }, + { + "epoch": 1.2730803996775217, + "grad_norm": 0.0011569177731871605, + "learning_rate": 0.08133520533502851, + "loss": 0.1132, + "num_input_tokens_seen": 33785216, + "step": 26055 + }, + { + "epoch": 1.2733247013412161, + "grad_norm": 0.001823537633754313, + "learning_rate": 0.08128283990386184, + "loss": 0.0993, + "num_input_tokens_seen": 33791584, + "step": 26060 + }, + { + "epoch": 1.2735690030049105, + "grad_norm": 0.0013509726850315928, + "learning_rate": 0.08123048506974488, + "loss": 0.0872, + "num_input_tokens_seen": 33797920, + "step": 26065 + }, + { + "epoch": 1.2738133046686049, + "grad_norm": 0.001683853450231254, + "learning_rate": 0.08117814084075124, + "loss": 0.0926, + "num_input_tokens_seen": 33804448, + "step": 26070 + }, + { + "epoch": 1.274057606332299, + "grad_norm": 0.0019777824636548758, + "learning_rate": 0.08112580722495318, + "loss": 0.092, + "num_input_tokens_seen": 33810976, + "step": 26075 + }, + { + "epoch": 1.2743019079959934, + "grad_norm": 0.0008974542724899948, + "learning_rate": 0.08107348423042122, + "loss": 0.0633, + "num_input_tokens_seen": 33817056, + "step": 26080 + }, + { + "epoch": 1.2745462096596878, + "grad_norm": 0.000560265441890806, + "learning_rate": 0.08102117186522413, + "loss": 0.0877, + "num_input_tokens_seen": 33823296, + "step": 26085 + }, + { + "epoch": 1.2747905113233822, + "grad_norm": 0.0018778059165924788, + "learning_rate": 0.08096887013742916, + "loss": 0.0737, + "num_input_tokens_seen": 33830048, + "step": 26090 + }, + { + "epoch": 1.2750348129870765, + "grad_norm": 0.0016863003838807344, + "learning_rate": 0.08091657905510198, + "loss": 0.0888, + "num_input_tokens_seen": 33836640, + "step": 26095 + }, + { + "epoch": 1.2752791146507707, + "grad_norm": 0.001415450475178659, + "learning_rate": 0.08086429862630642, + "loss": 0.088, + "num_input_tokens_seen": 33843328, + "step": 26100 + }, + { + "epoch": 1.275523416314465, + "grad_norm": 0.0012440497521311045, + "learning_rate": 0.08081202885910488, + "loss": 0.07, + "num_input_tokens_seen": 33850368, + "step": 26105 + }, + { + "epoch": 1.2757677179781595, + "grad_norm": 0.0013879077741876245, + "learning_rate": 0.08075976976155795, + "loss": 0.0987, + "num_input_tokens_seen": 33856480, + "step": 26110 + }, + { + "epoch": 1.2760120196418536, + "grad_norm": 0.0014043875271454453, + "learning_rate": 0.08070752134172461, + "loss": 0.0936, + "num_input_tokens_seen": 33862720, + "step": 26115 + }, + { + "epoch": 1.276256321305548, + "grad_norm": 0.0012200387427583337, + "learning_rate": 0.08065528360766229, + "loss": 0.0831, + "num_input_tokens_seen": 33869056, + "step": 26120 + }, + { + "epoch": 1.2765006229692424, + "grad_norm": 0.0009811237687245011, + "learning_rate": 0.08060305656742664, + "loss": 0.0898, + "num_input_tokens_seen": 33875488, + "step": 26125 + }, + { + "epoch": 1.2767449246329368, + "grad_norm": 0.000977837247774005, + "learning_rate": 0.08055084022907182, + "loss": 0.0988, + "num_input_tokens_seen": 33881920, + "step": 26130 + }, + { + "epoch": 1.2769892262966311, + "grad_norm": 0.0009669419378042221, + "learning_rate": 0.08049863460065014, + "loss": 0.0607, + "num_input_tokens_seen": 33888704, + "step": 26135 + }, + { + "epoch": 1.2772335279603255, + "grad_norm": 0.0022636603098362684, + "learning_rate": 0.0804464396902124, + "loss": 0.0833, + "num_input_tokens_seen": 33895328, + "step": 26140 + }, + { + "epoch": 1.2774778296240197, + "grad_norm": 0.0009922710014507174, + "learning_rate": 0.08039425550580777, + "loss": 0.0839, + "num_input_tokens_seen": 33901920, + "step": 26145 + }, + { + "epoch": 1.277722131287714, + "grad_norm": 0.0006864435854367912, + "learning_rate": 0.08034208205548363, + "loss": 0.0999, + "num_input_tokens_seen": 33908608, + "step": 26150 + }, + { + "epoch": 1.2779664329514084, + "grad_norm": 0.0016605586279183626, + "learning_rate": 0.08028991934728581, + "loss": 0.1243, + "num_input_tokens_seen": 33915072, + "step": 26155 + }, + { + "epoch": 1.2782107346151026, + "grad_norm": 0.001953283092007041, + "learning_rate": 0.0802377673892585, + "loss": 0.102, + "num_input_tokens_seen": 33921536, + "step": 26160 + }, + { + "epoch": 1.278455036278797, + "grad_norm": 0.0012008072808384895, + "learning_rate": 0.0801856261894441, + "loss": 0.1052, + "num_input_tokens_seen": 33927808, + "step": 26165 + }, + { + "epoch": 1.2786993379424914, + "grad_norm": 0.0013246263843029737, + "learning_rate": 0.08013349575588354, + "loss": 0.0929, + "num_input_tokens_seen": 33933856, + "step": 26170 + }, + { + "epoch": 1.2789436396061857, + "grad_norm": 0.0009185769013129175, + "learning_rate": 0.08008137609661586, + "loss": 0.0855, + "num_input_tokens_seen": 33940096, + "step": 26175 + }, + { + "epoch": 1.2791879412698801, + "grad_norm": 0.0009849794441834092, + "learning_rate": 0.08002926721967872, + "loss": 0.1008, + "num_input_tokens_seen": 33946528, + "step": 26180 + }, + { + "epoch": 1.2794322429335745, + "grad_norm": 0.0011661745375022292, + "learning_rate": 0.07997716913310782, + "loss": 0.0912, + "num_input_tokens_seen": 33952800, + "step": 26185 + }, + { + "epoch": 1.2796765445972687, + "grad_norm": 0.0022773013915866613, + "learning_rate": 0.07992508184493745, + "loss": 0.0912, + "num_input_tokens_seen": 33959136, + "step": 26190 + }, + { + "epoch": 1.279920846260963, + "grad_norm": 0.0014325878582894802, + "learning_rate": 0.07987300536320001, + "loss": 0.0863, + "num_input_tokens_seen": 33965888, + "step": 26195 + }, + { + "epoch": 1.2801651479246574, + "grad_norm": 0.001104680122807622, + "learning_rate": 0.07982093969592649, + "loss": 0.0885, + "num_input_tokens_seen": 33972576, + "step": 26200 + }, + { + "epoch": 1.2801651479246574, + "eval_loss": 0.09034081548452377, + "eval_runtime": 401.9047, + "eval_samples_per_second": 90.531, + "eval_steps_per_second": 22.635, + "num_input_tokens_seen": 33972576, + "step": 26200 + }, + { + "epoch": 1.2804094495883516, + "grad_norm": 0.0011314627481624484, + "learning_rate": 0.07976888485114592, + "loss": 0.0739, + "num_input_tokens_seen": 33978976, + "step": 26205 + }, + { + "epoch": 1.280653751252046, + "grad_norm": 0.0007173545309342444, + "learning_rate": 0.07971684083688595, + "loss": 0.0888, + "num_input_tokens_seen": 33985248, + "step": 26210 + }, + { + "epoch": 1.2808980529157403, + "grad_norm": 0.0006942196632735431, + "learning_rate": 0.0796648076611723, + "loss": 0.0819, + "num_input_tokens_seen": 33991264, + "step": 26215 + }, + { + "epoch": 1.2811423545794347, + "grad_norm": 0.0008080090046860278, + "learning_rate": 0.07961278533202922, + "loss": 0.0644, + "num_input_tokens_seen": 33997824, + "step": 26220 + }, + { + "epoch": 1.281386656243129, + "grad_norm": 0.0010628850432112813, + "learning_rate": 0.07956077385747919, + "loss": 0.0736, + "num_input_tokens_seen": 34004672, + "step": 26225 + }, + { + "epoch": 1.2816309579068235, + "grad_norm": 0.0005122520378790796, + "learning_rate": 0.079508773245543, + "loss": 0.0613, + "num_input_tokens_seen": 34011392, + "step": 26230 + }, + { + "epoch": 1.2818752595705176, + "grad_norm": 0.0010894244769588113, + "learning_rate": 0.07945678350423982, + "loss": 0.0952, + "num_input_tokens_seen": 34018016, + "step": 26235 + }, + { + "epoch": 1.282119561234212, + "grad_norm": 0.0014196099946275353, + "learning_rate": 0.07940480464158717, + "loss": 0.0686, + "num_input_tokens_seen": 34024192, + "step": 26240 + }, + { + "epoch": 1.2823638628979064, + "grad_norm": 0.000895873992703855, + "learning_rate": 0.07935283666560076, + "loss": 0.0799, + "num_input_tokens_seen": 34030496, + "step": 26245 + }, + { + "epoch": 1.2826081645616005, + "grad_norm": 0.0013302294537425041, + "learning_rate": 0.07930087958429478, + "loss": 0.099, + "num_input_tokens_seen": 34037632, + "step": 26250 + }, + { + "epoch": 1.282852466225295, + "grad_norm": 0.0015118059236556292, + "learning_rate": 0.07924893340568159, + "loss": 0.0879, + "num_input_tokens_seen": 34044256, + "step": 26255 + }, + { + "epoch": 1.2830967678889893, + "grad_norm": 0.0012138655874878168, + "learning_rate": 0.07919699813777205, + "loss": 0.093, + "num_input_tokens_seen": 34050752, + "step": 26260 + }, + { + "epoch": 1.2833410695526837, + "grad_norm": 0.0010044854134321213, + "learning_rate": 0.07914507378857515, + "loss": 0.0781, + "num_input_tokens_seen": 34057440, + "step": 26265 + }, + { + "epoch": 1.283585371216378, + "grad_norm": 0.0022276591043919325, + "learning_rate": 0.07909316036609822, + "loss": 0.0953, + "num_input_tokens_seen": 34063744, + "step": 26270 + }, + { + "epoch": 1.2838296728800724, + "grad_norm": 0.0013209165772423148, + "learning_rate": 0.07904125787834704, + "loss": 0.082, + "num_input_tokens_seen": 34069920, + "step": 26275 + }, + { + "epoch": 1.2840739745437666, + "grad_norm": 0.0010357270948588848, + "learning_rate": 0.07898936633332569, + "loss": 0.0702, + "num_input_tokens_seen": 34076224, + "step": 26280 + }, + { + "epoch": 1.284318276207461, + "grad_norm": 0.001572497421875596, + "learning_rate": 0.07893748573903635, + "loss": 0.0727, + "num_input_tokens_seen": 34082496, + "step": 26285 + }, + { + "epoch": 1.2845625778711554, + "grad_norm": 0.0010885950177907944, + "learning_rate": 0.0788856161034798, + "loss": 0.0914, + "num_input_tokens_seen": 34088992, + "step": 26290 + }, + { + "epoch": 1.2848068795348495, + "grad_norm": 0.001579725998453796, + "learning_rate": 0.07883375743465487, + "loss": 0.0674, + "num_input_tokens_seen": 34095296, + "step": 26295 + }, + { + "epoch": 1.285051181198544, + "grad_norm": 0.0016659063985571265, + "learning_rate": 0.07878190974055888, + "loss": 0.0858, + "num_input_tokens_seen": 34101632, + "step": 26300 + }, + { + "epoch": 1.2852954828622383, + "grad_norm": 0.0010919091291725636, + "learning_rate": 0.07873007302918746, + "loss": 0.0943, + "num_input_tokens_seen": 34108448, + "step": 26305 + }, + { + "epoch": 1.2855397845259326, + "grad_norm": 0.0016604174161329865, + "learning_rate": 0.07867824730853433, + "loss": 0.0792, + "num_input_tokens_seen": 34114816, + "step": 26310 + }, + { + "epoch": 1.285784086189627, + "grad_norm": 0.0010992665775120258, + "learning_rate": 0.07862643258659176, + "loss": 0.0795, + "num_input_tokens_seen": 34121408, + "step": 26315 + }, + { + "epoch": 1.2860283878533214, + "grad_norm": 0.001140510430559516, + "learning_rate": 0.07857462887135026, + "loss": 0.0763, + "num_input_tokens_seen": 34127776, + "step": 26320 + }, + { + "epoch": 1.2862726895170156, + "grad_norm": 0.0009916505077853799, + "learning_rate": 0.0785228361707986, + "loss": 0.0978, + "num_input_tokens_seen": 34134304, + "step": 26325 + }, + { + "epoch": 1.28651699118071, + "grad_norm": 0.0009621320059522986, + "learning_rate": 0.07847105449292378, + "loss": 0.0798, + "num_input_tokens_seen": 34140768, + "step": 26330 + }, + { + "epoch": 1.2867612928444043, + "grad_norm": 0.0017426683334633708, + "learning_rate": 0.0784192838457113, + "loss": 0.0964, + "num_input_tokens_seen": 34147200, + "step": 26335 + }, + { + "epoch": 1.2870055945080985, + "grad_norm": 0.0010877738241106272, + "learning_rate": 0.07836752423714473, + "loss": 0.083, + "num_input_tokens_seen": 34153664, + "step": 26340 + }, + { + "epoch": 1.2872498961717929, + "grad_norm": 0.0023746760562062263, + "learning_rate": 0.07831577567520616, + "loss": 0.0966, + "num_input_tokens_seen": 34159904, + "step": 26345 + }, + { + "epoch": 1.2874941978354872, + "grad_norm": 0.0010811175452545285, + "learning_rate": 0.07826403816787579, + "loss": 0.0688, + "num_input_tokens_seen": 34166400, + "step": 26350 + }, + { + "epoch": 1.2877384994991816, + "grad_norm": 0.0019301316933706403, + "learning_rate": 0.0782123117231322, + "loss": 0.0861, + "num_input_tokens_seen": 34173152, + "step": 26355 + }, + { + "epoch": 1.287982801162876, + "grad_norm": 0.0013060709461569786, + "learning_rate": 0.07816059634895237, + "loss": 0.0643, + "num_input_tokens_seen": 34179840, + "step": 26360 + }, + { + "epoch": 1.2882271028265704, + "grad_norm": 0.0011365690734237432, + "learning_rate": 0.0781088920533113, + "loss": 0.0858, + "num_input_tokens_seen": 34185824, + "step": 26365 + }, + { + "epoch": 1.2884714044902645, + "grad_norm": 0.0011940906988456845, + "learning_rate": 0.07805719884418257, + "loss": 0.1039, + "num_input_tokens_seen": 34192672, + "step": 26370 + }, + { + "epoch": 1.288715706153959, + "grad_norm": 0.0014948541065678, + "learning_rate": 0.07800551672953779, + "loss": 0.1092, + "num_input_tokens_seen": 34198816, + "step": 26375 + }, + { + "epoch": 1.2889600078176533, + "grad_norm": 0.0011701214825734496, + "learning_rate": 0.07795384571734709, + "loss": 0.154, + "num_input_tokens_seen": 34205888, + "step": 26380 + }, + { + "epoch": 1.2892043094813475, + "grad_norm": 0.0020676394924521446, + "learning_rate": 0.07790218581557883, + "loss": 0.1162, + "num_input_tokens_seen": 34212000, + "step": 26385 + }, + { + "epoch": 1.2894486111450418, + "grad_norm": 0.002524222247302532, + "learning_rate": 0.07785053703219949, + "loss": 0.0981, + "num_input_tokens_seen": 34218464, + "step": 26390 + }, + { + "epoch": 1.2896929128087362, + "grad_norm": 0.001360314548946917, + "learning_rate": 0.07779889937517409, + "loss": 0.0821, + "num_input_tokens_seen": 34224704, + "step": 26395 + }, + { + "epoch": 1.2899372144724306, + "grad_norm": 0.0013524782843887806, + "learning_rate": 0.0777472728524657, + "loss": 0.1014, + "num_input_tokens_seen": 34231488, + "step": 26400 + }, + { + "epoch": 1.2899372144724306, + "eval_loss": 0.08994686603546143, + "eval_runtime": 403.2432, + "eval_samples_per_second": 90.231, + "eval_steps_per_second": 22.56, + "num_input_tokens_seen": 34231488, + "step": 26400 + }, + { + "epoch": 1.290181516136125, + "grad_norm": 0.0010990308364853263, + "learning_rate": 0.07769565747203584, + "loss": 0.0912, + "num_input_tokens_seen": 34238016, + "step": 26405 + }, + { + "epoch": 1.2904258177998194, + "grad_norm": 0.00171353155747056, + "learning_rate": 0.07764405324184427, + "loss": 0.1174, + "num_input_tokens_seen": 34244416, + "step": 26410 + }, + { + "epoch": 1.2906701194635135, + "grad_norm": 0.000891003874130547, + "learning_rate": 0.07759246016984889, + "loss": 0.0866, + "num_input_tokens_seen": 34250560, + "step": 26415 + }, + { + "epoch": 1.290914421127208, + "grad_norm": 0.0014802063815295696, + "learning_rate": 0.07754087826400609, + "loss": 0.0834, + "num_input_tokens_seen": 34257504, + "step": 26420 + }, + { + "epoch": 1.2911587227909023, + "grad_norm": 0.001261106226593256, + "learning_rate": 0.0774893075322705, + "loss": 0.0883, + "num_input_tokens_seen": 34263776, + "step": 26425 + }, + { + "epoch": 1.2914030244545964, + "grad_norm": 0.0006982818595133722, + "learning_rate": 0.07743774798259484, + "loss": 0.076, + "num_input_tokens_seen": 34269792, + "step": 26430 + }, + { + "epoch": 1.2916473261182908, + "grad_norm": 0.0010491308057680726, + "learning_rate": 0.07738619962293032, + "loss": 0.0882, + "num_input_tokens_seen": 34276160, + "step": 26435 + }, + { + "epoch": 1.2918916277819852, + "grad_norm": 0.001493072253651917, + "learning_rate": 0.0773346624612264, + "loss": 0.0993, + "num_input_tokens_seen": 34282464, + "step": 26440 + }, + { + "epoch": 1.2921359294456796, + "grad_norm": 0.0008988179615698755, + "learning_rate": 0.07728313650543066, + "loss": 0.0859, + "num_input_tokens_seen": 34288512, + "step": 26445 + }, + { + "epoch": 1.292380231109374, + "grad_norm": 0.0011903827544301748, + "learning_rate": 0.07723162176348913, + "loss": 0.097, + "num_input_tokens_seen": 34294432, + "step": 26450 + }, + { + "epoch": 1.292624532773068, + "grad_norm": 0.0009165463270619512, + "learning_rate": 0.07718011824334593, + "loss": 0.0834, + "num_input_tokens_seen": 34300928, + "step": 26455 + }, + { + "epoch": 1.2928688344367625, + "grad_norm": 0.001313483458943665, + "learning_rate": 0.07712862595294363, + "loss": 0.0888, + "num_input_tokens_seen": 34307360, + "step": 26460 + }, + { + "epoch": 1.2931131361004569, + "grad_norm": 0.0010764694306999445, + "learning_rate": 0.07707714490022301, + "loss": 0.0891, + "num_input_tokens_seen": 34313792, + "step": 26465 + }, + { + "epoch": 1.2933574377641512, + "grad_norm": 0.001105735544115305, + "learning_rate": 0.07702567509312298, + "loss": 0.0985, + "num_input_tokens_seen": 34321664, + "step": 26470 + }, + { + "epoch": 1.2936017394278454, + "grad_norm": 0.0010878823231905699, + "learning_rate": 0.07697421653958098, + "loss": 0.0797, + "num_input_tokens_seen": 34328672, + "step": 26475 + }, + { + "epoch": 1.2938460410915398, + "grad_norm": 0.0022622912656515837, + "learning_rate": 0.07692276924753247, + "loss": 0.1027, + "num_input_tokens_seen": 34335232, + "step": 26480 + }, + { + "epoch": 1.2940903427552342, + "grad_norm": 0.0015783662674948573, + "learning_rate": 0.07687133322491124, + "loss": 0.1124, + "num_input_tokens_seen": 34342400, + "step": 26485 + }, + { + "epoch": 1.2943346444189285, + "grad_norm": 0.0012791563058272004, + "learning_rate": 0.07681990847964948, + "loss": 0.0772, + "num_input_tokens_seen": 34348736, + "step": 26490 + }, + { + "epoch": 1.294578946082623, + "grad_norm": 0.0008396261837333441, + "learning_rate": 0.0767684950196774, + "loss": 0.1033, + "num_input_tokens_seen": 34354976, + "step": 26495 + }, + { + "epoch": 1.294823247746317, + "grad_norm": 0.0014856290072202682, + "learning_rate": 0.0767170928529237, + "loss": 0.0641, + "num_input_tokens_seen": 34361472, + "step": 26500 + }, + { + "epoch": 1.2950675494100115, + "grad_norm": 0.0013744396856054664, + "learning_rate": 0.07666570198731526, + "loss": 0.0775, + "num_input_tokens_seen": 34367776, + "step": 26505 + }, + { + "epoch": 1.2953118510737058, + "grad_norm": 0.0008767107501626015, + "learning_rate": 0.07661432243077708, + "loss": 0.075, + "num_input_tokens_seen": 34375360, + "step": 26510 + }, + { + "epoch": 1.2955561527374002, + "grad_norm": 0.001246660714969039, + "learning_rate": 0.0765629541912326, + "loss": 0.0841, + "num_input_tokens_seen": 34381696, + "step": 26515 + }, + { + "epoch": 1.2958004544010944, + "grad_norm": 0.0012545052450150251, + "learning_rate": 0.07651159727660352, + "loss": 0.0858, + "num_input_tokens_seen": 34387968, + "step": 26520 + }, + { + "epoch": 1.2960447560647887, + "grad_norm": 0.001170053263194859, + "learning_rate": 0.07646025169480959, + "loss": 0.0697, + "num_input_tokens_seen": 34394464, + "step": 26525 + }, + { + "epoch": 1.2962890577284831, + "grad_norm": 0.001702590612694621, + "learning_rate": 0.07640891745376908, + "loss": 0.0943, + "num_input_tokens_seen": 34400608, + "step": 26530 + }, + { + "epoch": 1.2965333593921775, + "grad_norm": 0.0011353815207257867, + "learning_rate": 0.07635759456139822, + "loss": 0.0999, + "num_input_tokens_seen": 34406848, + "step": 26535 + }, + { + "epoch": 1.2967776610558719, + "grad_norm": 0.0018057749839499593, + "learning_rate": 0.0763062830256118, + "loss": 0.0936, + "num_input_tokens_seen": 34413664, + "step": 26540 + }, + { + "epoch": 1.297021962719566, + "grad_norm": 0.000995837152004242, + "learning_rate": 0.07625498285432258, + "loss": 0.0639, + "num_input_tokens_seen": 34420448, + "step": 26545 + }, + { + "epoch": 1.2972662643832604, + "grad_norm": 0.001653402461670339, + "learning_rate": 0.07620369405544176, + "loss": 0.0912, + "num_input_tokens_seen": 34427168, + "step": 26550 + }, + { + "epoch": 1.2975105660469548, + "grad_norm": 0.0008910009637475014, + "learning_rate": 0.07615241663687868, + "loss": 0.0807, + "num_input_tokens_seen": 34433824, + "step": 26555 + }, + { + "epoch": 1.2977548677106492, + "grad_norm": 0.002112144371494651, + "learning_rate": 0.07610115060654106, + "loss": 0.0851, + "num_input_tokens_seen": 34440480, + "step": 26560 + }, + { + "epoch": 1.2979991693743433, + "grad_norm": 0.0011157536646351218, + "learning_rate": 0.07604989597233458, + "loss": 0.0683, + "num_input_tokens_seen": 34447712, + "step": 26565 + }, + { + "epoch": 1.2982434710380377, + "grad_norm": 0.0010253711370751262, + "learning_rate": 0.07599865274216352, + "loss": 0.0883, + "num_input_tokens_seen": 34454560, + "step": 26570 + }, + { + "epoch": 1.298487772701732, + "grad_norm": 0.0011901299003511667, + "learning_rate": 0.07594742092393013, + "loss": 0.0897, + "num_input_tokens_seen": 34461216, + "step": 26575 + }, + { + "epoch": 1.2987320743654265, + "grad_norm": 0.0011055811773985624, + "learning_rate": 0.07589620052553503, + "loss": 0.0976, + "num_input_tokens_seen": 34467456, + "step": 26580 + }, + { + "epoch": 1.2989763760291209, + "grad_norm": 0.0009433326777070761, + "learning_rate": 0.0758449915548771, + "loss": 0.0821, + "num_input_tokens_seen": 34473504, + "step": 26585 + }, + { + "epoch": 1.299220677692815, + "grad_norm": 0.001317292801104486, + "learning_rate": 0.07579379401985332, + "loss": 0.0958, + "num_input_tokens_seen": 34479552, + "step": 26590 + }, + { + "epoch": 1.2994649793565094, + "grad_norm": 0.0009858320700004697, + "learning_rate": 0.07574260792835905, + "loss": 0.1044, + "num_input_tokens_seen": 34485568, + "step": 26595 + }, + { + "epoch": 1.2997092810202038, + "grad_norm": 0.0017173485830426216, + "learning_rate": 0.07569143328828784, + "loss": 0.106, + "num_input_tokens_seen": 34491904, + "step": 26600 + }, + { + "epoch": 1.2997092810202038, + "eval_loss": 0.09147468954324722, + "eval_runtime": 402.0229, + "eval_samples_per_second": 90.505, + "eval_steps_per_second": 22.628, + "num_input_tokens_seen": 34491904, + "step": 26600 + }, + { + "epoch": 1.2999535826838982, + "grad_norm": 0.0011748448014259338, + "learning_rate": 0.0756402701075314, + "loss": 0.0755, + "num_input_tokens_seen": 34498112, + "step": 26605 + }, + { + "epoch": 1.3001978843475923, + "grad_norm": 0.0010621696710586548, + "learning_rate": 0.07558911839397982, + "loss": 0.0877, + "num_input_tokens_seen": 34504256, + "step": 26610 + }, + { + "epoch": 1.3004421860112867, + "grad_norm": 0.0012653839075937867, + "learning_rate": 0.07553797815552123, + "loss": 0.0768, + "num_input_tokens_seen": 34510496, + "step": 26615 + }, + { + "epoch": 1.300686487674981, + "grad_norm": 0.0021941470913589, + "learning_rate": 0.07548684940004222, + "loss": 0.1131, + "num_input_tokens_seen": 34516672, + "step": 26620 + }, + { + "epoch": 1.3009307893386755, + "grad_norm": 0.0016921630594879389, + "learning_rate": 0.07543573213542744, + "loss": 0.1193, + "num_input_tokens_seen": 34522976, + "step": 26625 + }, + { + "epoch": 1.3011750910023698, + "grad_norm": 0.0005058835376985371, + "learning_rate": 0.0753846263695597, + "loss": 0.0794, + "num_input_tokens_seen": 34529632, + "step": 26630 + }, + { + "epoch": 1.301419392666064, + "grad_norm": 0.0014994359808042645, + "learning_rate": 0.07533353211032029, + "loss": 0.089, + "num_input_tokens_seen": 34536416, + "step": 26635 + }, + { + "epoch": 1.3016636943297584, + "grad_norm": 0.0006587738753296435, + "learning_rate": 0.07528244936558857, + "loss": 0.0684, + "num_input_tokens_seen": 34543136, + "step": 26640 + }, + { + "epoch": 1.3019079959934527, + "grad_norm": 0.001618970069102943, + "learning_rate": 0.07523137814324206, + "loss": 0.0954, + "num_input_tokens_seen": 34549504, + "step": 26645 + }, + { + "epoch": 1.302152297657147, + "grad_norm": 0.0010877352906391025, + "learning_rate": 0.07518031845115672, + "loss": 0.0757, + "num_input_tokens_seen": 34555840, + "step": 26650 + }, + { + "epoch": 1.3023965993208413, + "grad_norm": 0.0010902329813688993, + "learning_rate": 0.07512927029720647, + "loss": 0.0673, + "num_input_tokens_seen": 34563328, + "step": 26655 + }, + { + "epoch": 1.3026409009845357, + "grad_norm": 0.0006906437920406461, + "learning_rate": 0.0750782336892636, + "loss": 0.0706, + "num_input_tokens_seen": 34569600, + "step": 26660 + }, + { + "epoch": 1.30288520264823, + "grad_norm": 0.0013474604347720742, + "learning_rate": 0.0750272086351987, + "loss": 0.0914, + "num_input_tokens_seen": 34575712, + "step": 26665 + }, + { + "epoch": 1.3031295043119244, + "grad_norm": 0.0011208540527150035, + "learning_rate": 0.07497619514288031, + "loss": 0.092, + "num_input_tokens_seen": 34582272, + "step": 26670 + }, + { + "epoch": 1.3033738059756188, + "grad_norm": 0.0011706309160217643, + "learning_rate": 0.07492519322017545, + "loss": 0.0856, + "num_input_tokens_seen": 34588576, + "step": 26675 + }, + { + "epoch": 1.303618107639313, + "grad_norm": 0.0013145572738721967, + "learning_rate": 0.0748742028749493, + "loss": 0.0784, + "num_input_tokens_seen": 34595264, + "step": 26680 + }, + { + "epoch": 1.3038624093030073, + "grad_norm": 0.0006440733559429646, + "learning_rate": 0.0748232241150651, + "loss": 0.0708, + "num_input_tokens_seen": 34602144, + "step": 26685 + }, + { + "epoch": 1.3041067109667017, + "grad_norm": 0.00036820792593061924, + "learning_rate": 0.07477225694838453, + "loss": 0.0766, + "num_input_tokens_seen": 34608576, + "step": 26690 + }, + { + "epoch": 1.3043510126303959, + "grad_norm": 0.0008459951495751739, + "learning_rate": 0.07472130138276731, + "loss": 0.0882, + "num_input_tokens_seen": 34614880, + "step": 26695 + }, + { + "epoch": 1.3045953142940903, + "grad_norm": 0.00043714509229175746, + "learning_rate": 0.07467035742607138, + "loss": 0.097, + "num_input_tokens_seen": 34621536, + "step": 26700 + }, + { + "epoch": 1.3048396159577846, + "grad_norm": 0.001022462034597993, + "learning_rate": 0.07461942508615303, + "loss": 0.1047, + "num_input_tokens_seen": 34627776, + "step": 26705 + }, + { + "epoch": 1.305083917621479, + "grad_norm": 0.0009518723818473518, + "learning_rate": 0.07456850437086657, + "loss": 0.0909, + "num_input_tokens_seen": 34634272, + "step": 26710 + }, + { + "epoch": 1.3053282192851734, + "grad_norm": 0.0017677476862445474, + "learning_rate": 0.07451759528806468, + "loss": 0.1053, + "num_input_tokens_seen": 34640960, + "step": 26715 + }, + { + "epoch": 1.3055725209488678, + "grad_norm": 0.0008658102597109973, + "learning_rate": 0.0744666978455982, + "loss": 0.0745, + "num_input_tokens_seen": 34647296, + "step": 26720 + }, + { + "epoch": 1.305816822612562, + "grad_norm": 0.0012175773736089468, + "learning_rate": 0.07441581205131609, + "loss": 0.0873, + "num_input_tokens_seen": 34653920, + "step": 26725 + }, + { + "epoch": 1.3060611242762563, + "grad_norm": 0.0008248942904174328, + "learning_rate": 0.07436493791306566, + "loss": 0.0907, + "num_input_tokens_seen": 34660224, + "step": 26730 + }, + { + "epoch": 1.3063054259399507, + "grad_norm": 0.0012454040115699172, + "learning_rate": 0.07431407543869223, + "loss": 0.0943, + "num_input_tokens_seen": 34666784, + "step": 26735 + }, + { + "epoch": 1.3065497276036449, + "grad_norm": 0.001479150028899312, + "learning_rate": 0.0742632246360395, + "loss": 0.0845, + "num_input_tokens_seen": 34673760, + "step": 26740 + }, + { + "epoch": 1.3067940292673392, + "grad_norm": 0.002082020975649357, + "learning_rate": 0.07421238551294934, + "loss": 0.0954, + "num_input_tokens_seen": 34679968, + "step": 26745 + }, + { + "epoch": 1.3070383309310336, + "grad_norm": 0.0014080768451094627, + "learning_rate": 0.07416155807726171, + "loss": 0.0842, + "num_input_tokens_seen": 34686560, + "step": 26750 + }, + { + "epoch": 1.307282632594728, + "grad_norm": 0.001298877876251936, + "learning_rate": 0.07411074233681492, + "loss": 0.0925, + "num_input_tokens_seen": 34692864, + "step": 26755 + }, + { + "epoch": 1.3075269342584224, + "grad_norm": 0.001248466200195253, + "learning_rate": 0.07405993829944528, + "loss": 0.0935, + "num_input_tokens_seen": 34699040, + "step": 26760 + }, + { + "epoch": 1.3077712359221167, + "grad_norm": 0.000721468823030591, + "learning_rate": 0.07400914597298755, + "loss": 0.0657, + "num_input_tokens_seen": 34705248, + "step": 26765 + }, + { + "epoch": 1.308015537585811, + "grad_norm": 0.001496599754318595, + "learning_rate": 0.07395836536527445, + "loss": 0.1036, + "num_input_tokens_seen": 34711904, + "step": 26770 + }, + { + "epoch": 1.3082598392495053, + "grad_norm": 0.0006484561017714441, + "learning_rate": 0.07390759648413696, + "loss": 0.0828, + "num_input_tokens_seen": 34718272, + "step": 26775 + }, + { + "epoch": 1.3085041409131997, + "grad_norm": 0.0008125486201606691, + "learning_rate": 0.07385683933740435, + "loss": 0.0757, + "num_input_tokens_seen": 34724768, + "step": 26780 + }, + { + "epoch": 1.3087484425768938, + "grad_norm": 0.0006416343967430294, + "learning_rate": 0.07380609393290402, + "loss": 0.0894, + "num_input_tokens_seen": 34731232, + "step": 26785 + }, + { + "epoch": 1.3089927442405882, + "grad_norm": 0.001301950542256236, + "learning_rate": 0.07375536027846147, + "loss": 0.0732, + "num_input_tokens_seen": 34737984, + "step": 26790 + }, + { + "epoch": 1.3092370459042826, + "grad_norm": 0.0017812781734392047, + "learning_rate": 0.07370463838190057, + "loss": 0.1097, + "num_input_tokens_seen": 34744480, + "step": 26795 + }, + { + "epoch": 1.309481347567977, + "grad_norm": 0.000874718592967838, + "learning_rate": 0.07365392825104317, + "loss": 0.076, + "num_input_tokens_seen": 34751008, + "step": 26800 + }, + { + "epoch": 1.309481347567977, + "eval_loss": 0.08869751542806625, + "eval_runtime": 402.4415, + "eval_samples_per_second": 90.411, + "eval_steps_per_second": 22.605, + "num_input_tokens_seen": 34751008, + "step": 26800 + }, + { + "epoch": 1.3097256492316713, + "grad_norm": 0.0014464680571109056, + "learning_rate": 0.07360322989370945, + "loss": 0.0779, + "num_input_tokens_seen": 34757440, + "step": 26805 + }, + { + "epoch": 1.3099699508953657, + "grad_norm": 0.0021597028244286776, + "learning_rate": 0.07355254331771781, + "loss": 0.0846, + "num_input_tokens_seen": 34763840, + "step": 26810 + }, + { + "epoch": 1.3102142525590599, + "grad_norm": 0.001938295317813754, + "learning_rate": 0.07350186853088461, + "loss": 0.0896, + "num_input_tokens_seen": 34770176, + "step": 26815 + }, + { + "epoch": 1.3104585542227543, + "grad_norm": 0.002741581527516246, + "learning_rate": 0.07345120554102462, + "loss": 0.0928, + "num_input_tokens_seen": 34776576, + "step": 26820 + }, + { + "epoch": 1.3107028558864486, + "grad_norm": 0.0030454122461378574, + "learning_rate": 0.07340055435595079, + "loss": 0.0919, + "num_input_tokens_seen": 34783136, + "step": 26825 + }, + { + "epoch": 1.3109471575501428, + "grad_norm": 0.0010632039047777653, + "learning_rate": 0.07334991498347401, + "loss": 0.0856, + "num_input_tokens_seen": 34789344, + "step": 26830 + }, + { + "epoch": 1.3111914592138372, + "grad_norm": 0.0010401956969872117, + "learning_rate": 0.07329928743140365, + "loss": 0.0861, + "num_input_tokens_seen": 34795584, + "step": 26835 + }, + { + "epoch": 1.3114357608775316, + "grad_norm": 0.0007470008567906916, + "learning_rate": 0.07324867170754705, + "loss": 0.0834, + "num_input_tokens_seen": 34801600, + "step": 26840 + }, + { + "epoch": 1.311680062541226, + "grad_norm": 0.001017797039821744, + "learning_rate": 0.07319806781970974, + "loss": 0.0923, + "num_input_tokens_seen": 34807456, + "step": 26845 + }, + { + "epoch": 1.3119243642049203, + "grad_norm": 0.0011153430677950382, + "learning_rate": 0.07314747577569555, + "loss": 0.0762, + "num_input_tokens_seen": 34814240, + "step": 26850 + }, + { + "epoch": 1.3121686658686147, + "grad_norm": 0.0020157862454652786, + "learning_rate": 0.07309689558330636, + "loss": 0.0985, + "num_input_tokens_seen": 34820416, + "step": 26855 + }, + { + "epoch": 1.3124129675323088, + "grad_norm": 0.00133793824352324, + "learning_rate": 0.0730463272503423, + "loss": 0.071, + "num_input_tokens_seen": 34827936, + "step": 26860 + }, + { + "epoch": 1.3126572691960032, + "grad_norm": 0.0013680533738806844, + "learning_rate": 0.07299577078460168, + "loss": 0.1141, + "num_input_tokens_seen": 34833856, + "step": 26865 + }, + { + "epoch": 1.3129015708596976, + "grad_norm": 0.0012766581494361162, + "learning_rate": 0.07294522619388083, + "loss": 0.0707, + "num_input_tokens_seen": 34840672, + "step": 26870 + }, + { + "epoch": 1.3131458725233918, + "grad_norm": 0.0014528827741742134, + "learning_rate": 0.07289469348597452, + "loss": 0.0854, + "num_input_tokens_seen": 34847872, + "step": 26875 + }, + { + "epoch": 1.3133901741870861, + "grad_norm": 0.0019050101982429624, + "learning_rate": 0.07284417266867535, + "loss": 0.0835, + "num_input_tokens_seen": 34854176, + "step": 26880 + }, + { + "epoch": 1.3136344758507805, + "grad_norm": 0.0013080431381240487, + "learning_rate": 0.07279366374977439, + "loss": 0.1037, + "num_input_tokens_seen": 34860000, + "step": 26885 + }, + { + "epoch": 1.313878777514475, + "grad_norm": 0.0023010633885860443, + "learning_rate": 0.07274316673706074, + "loss": 0.0965, + "num_input_tokens_seen": 34866560, + "step": 26890 + }, + { + "epoch": 1.3141230791781693, + "grad_norm": 0.0016631537582725286, + "learning_rate": 0.07269268163832161, + "loss": 0.0836, + "num_input_tokens_seen": 34872672, + "step": 26895 + }, + { + "epoch": 1.3143673808418637, + "grad_norm": 0.0015951954992488027, + "learning_rate": 0.07264220846134248, + "loss": 0.0997, + "num_input_tokens_seen": 34878848, + "step": 26900 + }, + { + "epoch": 1.3146116825055578, + "grad_norm": 0.0011363248340785503, + "learning_rate": 0.07259174721390699, + "loss": 0.0685, + "num_input_tokens_seen": 34885664, + "step": 26905 + }, + { + "epoch": 1.3148559841692522, + "grad_norm": 0.001540876692160964, + "learning_rate": 0.07254129790379686, + "loss": 0.1122, + "num_input_tokens_seen": 34892032, + "step": 26910 + }, + { + "epoch": 1.3151002858329466, + "grad_norm": 0.0012400284176692367, + "learning_rate": 0.072490860538792, + "loss": 0.0797, + "num_input_tokens_seen": 34898272, + "step": 26915 + }, + { + "epoch": 1.3153445874966407, + "grad_norm": 0.0009272955358028412, + "learning_rate": 0.07244043512667042, + "loss": 0.0791, + "num_input_tokens_seen": 34904448, + "step": 26920 + }, + { + "epoch": 1.3155888891603351, + "grad_norm": 0.0028412560932338238, + "learning_rate": 0.07239002167520843, + "loss": 0.0931, + "num_input_tokens_seen": 34911104, + "step": 26925 + }, + { + "epoch": 1.3158331908240295, + "grad_norm": 0.0009385824669152498, + "learning_rate": 0.07233962019218045, + "loss": 0.0567, + "num_input_tokens_seen": 34917568, + "step": 26930 + }, + { + "epoch": 1.3160774924877239, + "grad_norm": 0.001289862091653049, + "learning_rate": 0.07228923068535892, + "loss": 0.0937, + "num_input_tokens_seen": 34924000, + "step": 26935 + }, + { + "epoch": 1.3163217941514183, + "grad_norm": 0.001517180586233735, + "learning_rate": 0.0722388531625146, + "loss": 0.0766, + "num_input_tokens_seen": 34930816, + "step": 26940 + }, + { + "epoch": 1.3165660958151126, + "grad_norm": 0.0008982345461845398, + "learning_rate": 0.07218848763141639, + "loss": 0.068, + "num_input_tokens_seen": 34937312, + "step": 26945 + }, + { + "epoch": 1.3168103974788068, + "grad_norm": 0.00082999630831182, + "learning_rate": 0.07213813409983118, + "loss": 0.087, + "num_input_tokens_seen": 34943200, + "step": 26950 + }, + { + "epoch": 1.3170546991425012, + "grad_norm": 0.0011225645430386066, + "learning_rate": 0.0720877925755242, + "loss": 0.0814, + "num_input_tokens_seen": 34949408, + "step": 26955 + }, + { + "epoch": 1.3172990008061956, + "grad_norm": 0.0017355959862470627, + "learning_rate": 0.07203746306625866, + "loss": 0.0893, + "num_input_tokens_seen": 34955616, + "step": 26960 + }, + { + "epoch": 1.3175433024698897, + "grad_norm": 0.0012007992481812835, + "learning_rate": 0.07198714557979606, + "loss": 0.0876, + "num_input_tokens_seen": 34961888, + "step": 26965 + }, + { + "epoch": 1.317787604133584, + "grad_norm": 0.00090679811546579, + "learning_rate": 0.07193684012389602, + "loss": 0.0877, + "num_input_tokens_seen": 34968352, + "step": 26970 + }, + { + "epoch": 1.3180319057972785, + "grad_norm": 0.001580013195052743, + "learning_rate": 0.07188654670631621, + "loss": 0.1029, + "num_input_tokens_seen": 34974624, + "step": 26975 + }, + { + "epoch": 1.3182762074609728, + "grad_norm": 0.0011883812258020043, + "learning_rate": 0.07183626533481258, + "loss": 0.1278, + "num_input_tokens_seen": 34980576, + "step": 26980 + }, + { + "epoch": 1.3185205091246672, + "grad_norm": 0.001101488247513771, + "learning_rate": 0.07178599601713909, + "loss": 0.0755, + "num_input_tokens_seen": 34987040, + "step": 26985 + }, + { + "epoch": 1.3187648107883614, + "grad_norm": 0.0010036792373284698, + "learning_rate": 0.07173573876104786, + "loss": 0.1147, + "num_input_tokens_seen": 34993856, + "step": 26990 + }, + { + "epoch": 1.3190091124520558, + "grad_norm": 0.001516419812105596, + "learning_rate": 0.0716854935742893, + "loss": 0.0763, + "num_input_tokens_seen": 35000000, + "step": 26995 + }, + { + "epoch": 1.3192534141157501, + "grad_norm": 0.0013793669641017914, + "learning_rate": 0.07163526046461174, + "loss": 0.0884, + "num_input_tokens_seen": 35006432, + "step": 27000 + }, + { + "epoch": 1.3192534141157501, + "eval_loss": 0.09002908319234848, + "eval_runtime": 402.3909, + "eval_samples_per_second": 90.422, + "eval_steps_per_second": 22.607, + "num_input_tokens_seen": 35006432, + "step": 27000 + }, + { + "epoch": 1.3194977157794445, + "grad_norm": 0.0014493736671283841, + "learning_rate": 0.07158503943976181, + "loss": 0.0909, + "num_input_tokens_seen": 35012608, + "step": 27005 + }, + { + "epoch": 1.3197420174431387, + "grad_norm": 0.0015992078697308898, + "learning_rate": 0.07153483050748427, + "loss": 0.089, + "num_input_tokens_seen": 35019136, + "step": 27010 + }, + { + "epoch": 1.319986319106833, + "grad_norm": 0.0014035971835255623, + "learning_rate": 0.07148463367552188, + "loss": 0.0833, + "num_input_tokens_seen": 35025600, + "step": 27015 + }, + { + "epoch": 1.3202306207705274, + "grad_norm": 0.0011651759268715978, + "learning_rate": 0.07143444895161565, + "loss": 0.0782, + "num_input_tokens_seen": 35031936, + "step": 27020 + }, + { + "epoch": 1.3204749224342218, + "grad_norm": 0.0011298226891085505, + "learning_rate": 0.07138427634350476, + "loss": 0.0828, + "num_input_tokens_seen": 35038592, + "step": 27025 + }, + { + "epoch": 1.3207192240979162, + "grad_norm": 0.0012721898965537548, + "learning_rate": 0.07133411585892636, + "loss": 0.0541, + "num_input_tokens_seen": 35045216, + "step": 27030 + }, + { + "epoch": 1.3209635257616104, + "grad_norm": 0.001091091544367373, + "learning_rate": 0.07128396750561593, + "loss": 0.0947, + "num_input_tokens_seen": 35051744, + "step": 27035 + }, + { + "epoch": 1.3212078274253047, + "grad_norm": 0.00111380813177675, + "learning_rate": 0.07123383129130685, + "loss": 0.0906, + "num_input_tokens_seen": 35058304, + "step": 27040 + }, + { + "epoch": 1.3214521290889991, + "grad_norm": 0.0011262116022408009, + "learning_rate": 0.07118370722373084, + "loss": 0.0991, + "num_input_tokens_seen": 35065056, + "step": 27045 + }, + { + "epoch": 1.3216964307526935, + "grad_norm": 0.001292427652515471, + "learning_rate": 0.07113359531061769, + "loss": 0.0639, + "num_input_tokens_seen": 35072288, + "step": 27050 + }, + { + "epoch": 1.3219407324163877, + "grad_norm": 0.001325528952293098, + "learning_rate": 0.07108349555969525, + "loss": 0.0755, + "num_input_tokens_seen": 35078752, + "step": 27055 + }, + { + "epoch": 1.322185034080082, + "grad_norm": 0.0010754039976745844, + "learning_rate": 0.07103340797868944, + "loss": 0.0914, + "num_input_tokens_seen": 35084992, + "step": 27060 + }, + { + "epoch": 1.3224293357437764, + "grad_norm": 0.0012021985603496432, + "learning_rate": 0.07098333257532453, + "loss": 0.0833, + "num_input_tokens_seen": 35091584, + "step": 27065 + }, + { + "epoch": 1.3226736374074708, + "grad_norm": 0.0017539025284349918, + "learning_rate": 0.07093326935732269, + "loss": 0.0878, + "num_input_tokens_seen": 35097504, + "step": 27070 + }, + { + "epoch": 1.3229179390711652, + "grad_norm": 0.0015372791094705462, + "learning_rate": 0.0708832183324044, + "loss": 0.1035, + "num_input_tokens_seen": 35104032, + "step": 27075 + }, + { + "epoch": 1.3231622407348593, + "grad_norm": 0.001103056943975389, + "learning_rate": 0.07083317950828799, + "loss": 0.0676, + "num_input_tokens_seen": 35110560, + "step": 27080 + }, + { + "epoch": 1.3234065423985537, + "grad_norm": 0.0018895026296377182, + "learning_rate": 0.0707831528926902, + "loss": 0.113, + "num_input_tokens_seen": 35116576, + "step": 27085 + }, + { + "epoch": 1.323650844062248, + "grad_norm": 0.0016137759666889906, + "learning_rate": 0.07073313849332578, + "loss": 0.0989, + "num_input_tokens_seen": 35122880, + "step": 27090 + }, + { + "epoch": 1.3238951457259425, + "grad_norm": 0.0011104138102382421, + "learning_rate": 0.07068313631790749, + "loss": 0.0964, + "num_input_tokens_seen": 35130016, + "step": 27095 + }, + { + "epoch": 1.3241394473896366, + "grad_norm": 0.0012233442394062877, + "learning_rate": 0.07063314637414632, + "loss": 0.1104, + "num_input_tokens_seen": 35136256, + "step": 27100 + }, + { + "epoch": 1.324383749053331, + "grad_norm": 0.001104273833334446, + "learning_rate": 0.07058316866975144, + "loss": 0.075, + "num_input_tokens_seen": 35142656, + "step": 27105 + }, + { + "epoch": 1.3246280507170254, + "grad_norm": 0.0009551111725158989, + "learning_rate": 0.0705332032124299, + "loss": 0.0634, + "num_input_tokens_seen": 35148864, + "step": 27110 + }, + { + "epoch": 1.3248723523807198, + "grad_norm": 0.000742158736102283, + "learning_rate": 0.0704832500098871, + "loss": 0.0816, + "num_input_tokens_seen": 35155360, + "step": 27115 + }, + { + "epoch": 1.3251166540444141, + "grad_norm": 0.001761500257998705, + "learning_rate": 0.07043330906982641, + "loss": 0.1094, + "num_input_tokens_seen": 35161376, + "step": 27120 + }, + { + "epoch": 1.3253609557081083, + "grad_norm": 0.0015003994340077043, + "learning_rate": 0.07038338039994936, + "loss": 0.1064, + "num_input_tokens_seen": 35167840, + "step": 27125 + }, + { + "epoch": 1.3256052573718027, + "grad_norm": 0.0012235598405823112, + "learning_rate": 0.07033346400795562, + "loss": 0.0712, + "num_input_tokens_seen": 35174944, + "step": 27130 + }, + { + "epoch": 1.325849559035497, + "grad_norm": 0.0013609741581603885, + "learning_rate": 0.07028355990154282, + "loss": 0.0767, + "num_input_tokens_seen": 35180928, + "step": 27135 + }, + { + "epoch": 1.3260938606991914, + "grad_norm": 0.001877589849755168, + "learning_rate": 0.07023366808840685, + "loss": 0.1051, + "num_input_tokens_seen": 35187424, + "step": 27140 + }, + { + "epoch": 1.3263381623628856, + "grad_norm": 0.0012434180825948715, + "learning_rate": 0.07018378857624172, + "loss": 0.072, + "num_input_tokens_seen": 35193792, + "step": 27145 + }, + { + "epoch": 1.32658246402658, + "grad_norm": 0.0010081867221742868, + "learning_rate": 0.0701339213727394, + "loss": 0.0693, + "num_input_tokens_seen": 35199936, + "step": 27150 + }, + { + "epoch": 1.3268267656902744, + "grad_norm": 0.0012542795157060027, + "learning_rate": 0.07008406648559008, + "loss": 0.0891, + "num_input_tokens_seen": 35206304, + "step": 27155 + }, + { + "epoch": 1.3270710673539687, + "grad_norm": 0.0012623437214642763, + "learning_rate": 0.07003422392248196, + "loss": 0.0876, + "num_input_tokens_seen": 35212544, + "step": 27160 + }, + { + "epoch": 1.3273153690176631, + "grad_norm": 0.0017139245755970478, + "learning_rate": 0.06998439369110142, + "loss": 0.0885, + "num_input_tokens_seen": 35219616, + "step": 27165 + }, + { + "epoch": 1.3275596706813573, + "grad_norm": 0.0012557992013171315, + "learning_rate": 0.06993457579913295, + "loss": 0.0819, + "num_input_tokens_seen": 35225984, + "step": 27170 + }, + { + "epoch": 1.3278039723450517, + "grad_norm": 0.0011062187841162086, + "learning_rate": 0.06988477025425903, + "loss": 0.1042, + "num_input_tokens_seen": 35232672, + "step": 27175 + }, + { + "epoch": 1.328048274008746, + "grad_norm": 0.001190823270007968, + "learning_rate": 0.06983497706416032, + "loss": 0.0796, + "num_input_tokens_seen": 35239072, + "step": 27180 + }, + { + "epoch": 1.3282925756724404, + "grad_norm": 0.0011805641697719693, + "learning_rate": 0.0697851962365156, + "loss": 0.0943, + "num_input_tokens_seen": 35245376, + "step": 27185 + }, + { + "epoch": 1.3285368773361346, + "grad_norm": 0.0006492737447842956, + "learning_rate": 0.06973542777900163, + "loss": 0.0853, + "num_input_tokens_seen": 35251968, + "step": 27190 + }, + { + "epoch": 1.328781178999829, + "grad_norm": 0.0009159549372270703, + "learning_rate": 0.06968567169929342, + "loss": 0.0827, + "num_input_tokens_seen": 35258624, + "step": 27195 + }, + { + "epoch": 1.3290254806635233, + "grad_norm": 0.0010959157953038812, + "learning_rate": 0.06963592800506392, + "loss": 0.0715, + "num_input_tokens_seen": 35264896, + "step": 27200 + }, + { + "epoch": 1.3290254806635233, + "eval_loss": 0.09126756340265274, + "eval_runtime": 402.5556, + "eval_samples_per_second": 90.385, + "eval_steps_per_second": 22.598, + "num_input_tokens_seen": 35264896, + "step": 27200 + }, + { + "epoch": 1.3292697823272177, + "grad_norm": 0.0010134275071322918, + "learning_rate": 0.06958619670398417, + "loss": 0.1021, + "num_input_tokens_seen": 35271040, + "step": 27205 + }, + { + "epoch": 1.329514083990912, + "grad_norm": 0.0011670882813632488, + "learning_rate": 0.0695364778037235, + "loss": 0.1047, + "num_input_tokens_seen": 35277472, + "step": 27210 + }, + { + "epoch": 1.3297583856546062, + "grad_norm": 0.001227181521244347, + "learning_rate": 0.06948677131194907, + "loss": 0.0876, + "num_input_tokens_seen": 35284384, + "step": 27215 + }, + { + "epoch": 1.3300026873183006, + "grad_norm": 0.0013625134015455842, + "learning_rate": 0.06943707723632629, + "loss": 0.1084, + "num_input_tokens_seen": 35290944, + "step": 27220 + }, + { + "epoch": 1.330246988981995, + "grad_norm": 0.0009955349378287792, + "learning_rate": 0.06938739558451867, + "loss": 0.1067, + "num_input_tokens_seen": 35297312, + "step": 27225 + }, + { + "epoch": 1.3304912906456892, + "grad_norm": 0.0010926579125225544, + "learning_rate": 0.06933772636418763, + "loss": 0.0909, + "num_input_tokens_seen": 35303520, + "step": 27230 + }, + { + "epoch": 1.3307355923093835, + "grad_norm": 0.00105603679548949, + "learning_rate": 0.06928806958299293, + "loss": 0.0809, + "num_input_tokens_seen": 35309472, + "step": 27235 + }, + { + "epoch": 1.330979893973078, + "grad_norm": 0.0015358718810603023, + "learning_rate": 0.06923842524859211, + "loss": 0.0732, + "num_input_tokens_seen": 35315584, + "step": 27240 + }, + { + "epoch": 1.3312241956367723, + "grad_norm": 0.0012133611598983407, + "learning_rate": 0.06918879336864105, + "loss": 0.0653, + "num_input_tokens_seen": 35321472, + "step": 27245 + }, + { + "epoch": 1.3314684973004667, + "grad_norm": 0.0016743679298087955, + "learning_rate": 0.06913917395079362, + "loss": 0.0969, + "num_input_tokens_seen": 35328000, + "step": 27250 + }, + { + "epoch": 1.331712798964161, + "grad_norm": 0.0012414120137691498, + "learning_rate": 0.0690895670027017, + "loss": 0.0858, + "num_input_tokens_seen": 35334368, + "step": 27255 + }, + { + "epoch": 1.3319571006278552, + "grad_norm": 0.002027191687375307, + "learning_rate": 0.06903997253201531, + "loss": 0.1043, + "num_input_tokens_seen": 35341024, + "step": 27260 + }, + { + "epoch": 1.3322014022915496, + "grad_norm": 0.0011270925169810653, + "learning_rate": 0.06899039054638263, + "loss": 0.0661, + "num_input_tokens_seen": 35347296, + "step": 27265 + }, + { + "epoch": 1.332445703955244, + "grad_norm": 0.0011137448018416762, + "learning_rate": 0.06894082105344976, + "loss": 0.0786, + "num_input_tokens_seen": 35353632, + "step": 27270 + }, + { + "epoch": 1.3326900056189381, + "grad_norm": 0.0020361319184303284, + "learning_rate": 0.06889126406086087, + "loss": 0.097, + "num_input_tokens_seen": 35360192, + "step": 27275 + }, + { + "epoch": 1.3329343072826325, + "grad_norm": 0.0018720004009082913, + "learning_rate": 0.0688417195762584, + "loss": 0.079, + "num_input_tokens_seen": 35366816, + "step": 27280 + }, + { + "epoch": 1.333178608946327, + "grad_norm": 0.0012084882473573089, + "learning_rate": 0.06879218760728262, + "loss": 0.0807, + "num_input_tokens_seen": 35373280, + "step": 27285 + }, + { + "epoch": 1.3334229106100213, + "grad_norm": 0.0015646048123016953, + "learning_rate": 0.06874266816157207, + "loss": 0.0796, + "num_input_tokens_seen": 35379648, + "step": 27290 + }, + { + "epoch": 1.3336672122737157, + "grad_norm": 0.0010090904543176293, + "learning_rate": 0.06869316124676321, + "loss": 0.0901, + "num_input_tokens_seen": 35385952, + "step": 27295 + }, + { + "epoch": 1.33391151393741, + "grad_norm": 0.0006766310543753207, + "learning_rate": 0.06864366687049062, + "loss": 0.0605, + "num_input_tokens_seen": 35392576, + "step": 27300 + }, + { + "epoch": 1.3341558156011042, + "grad_norm": 0.0012636922765523195, + "learning_rate": 0.06859418504038704, + "loss": 0.0674, + "num_input_tokens_seen": 35399584, + "step": 27305 + }, + { + "epoch": 1.3344001172647986, + "grad_norm": 0.0011781689245253801, + "learning_rate": 0.06854471576408311, + "loss": 0.0856, + "num_input_tokens_seen": 35406432, + "step": 27310 + }, + { + "epoch": 1.334644418928493, + "grad_norm": 0.0013499725610017776, + "learning_rate": 0.06849525904920767, + "loss": 0.087, + "num_input_tokens_seen": 35412768, + "step": 27315 + }, + { + "epoch": 1.334888720592187, + "grad_norm": 0.0008775359601713717, + "learning_rate": 0.06844581490338748, + "loss": 0.0917, + "num_input_tokens_seen": 35419104, + "step": 27320 + }, + { + "epoch": 1.3351330222558815, + "grad_norm": 0.0011442144168540835, + "learning_rate": 0.06839638333424752, + "loss": 0.0642, + "num_input_tokens_seen": 35425824, + "step": 27325 + }, + { + "epoch": 1.3353773239195759, + "grad_norm": 0.0013269014889374375, + "learning_rate": 0.06834696434941082, + "loss": 0.0848, + "num_input_tokens_seen": 35432256, + "step": 27330 + }, + { + "epoch": 1.3356216255832702, + "grad_norm": 0.0013370220549404621, + "learning_rate": 0.06829755795649824, + "loss": 0.0643, + "num_input_tokens_seen": 35439040, + "step": 27335 + }, + { + "epoch": 1.3358659272469646, + "grad_norm": 0.0017985805170610547, + "learning_rate": 0.06824816416312904, + "loss": 0.0927, + "num_input_tokens_seen": 35445280, + "step": 27340 + }, + { + "epoch": 1.336110228910659, + "grad_norm": 0.002464412711560726, + "learning_rate": 0.06819878297692027, + "loss": 0.0909, + "num_input_tokens_seen": 35451872, + "step": 27345 + }, + { + "epoch": 1.3363545305743532, + "grad_norm": 0.0014870903687551618, + "learning_rate": 0.0681494144054871, + "loss": 0.1355, + "num_input_tokens_seen": 35458240, + "step": 27350 + }, + { + "epoch": 1.3365988322380475, + "grad_norm": 0.0011595955584198236, + "learning_rate": 0.06810005845644286, + "loss": 0.0853, + "num_input_tokens_seen": 35464640, + "step": 27355 + }, + { + "epoch": 1.336843133901742, + "grad_norm": 0.0009574935538694263, + "learning_rate": 0.06805071513739878, + "loss": 0.0799, + "num_input_tokens_seen": 35471456, + "step": 27360 + }, + { + "epoch": 1.337087435565436, + "grad_norm": 0.0010496145114302635, + "learning_rate": 0.06800138445596428, + "loss": 0.0683, + "num_input_tokens_seen": 35477792, + "step": 27365 + }, + { + "epoch": 1.3373317372291305, + "grad_norm": 0.0012296028435230255, + "learning_rate": 0.06795206641974678, + "loss": 0.0776, + "num_input_tokens_seen": 35484000, + "step": 27370 + }, + { + "epoch": 1.3375760388928248, + "grad_norm": 0.00115953478962183, + "learning_rate": 0.06790276103635169, + "loss": 0.0785, + "num_input_tokens_seen": 35490752, + "step": 27375 + }, + { + "epoch": 1.3378203405565192, + "grad_norm": 0.0008042444242164493, + "learning_rate": 0.0678534683133826, + "loss": 0.0676, + "num_input_tokens_seen": 35497248, + "step": 27380 + }, + { + "epoch": 1.3380646422202136, + "grad_norm": 0.0008884879644028842, + "learning_rate": 0.06780418825844095, + "loss": 0.0723, + "num_input_tokens_seen": 35504000, + "step": 27385 + }, + { + "epoch": 1.338308943883908, + "grad_norm": 0.0010169543093070388, + "learning_rate": 0.0677549208791264, + "loss": 0.0746, + "num_input_tokens_seen": 35510688, + "step": 27390 + }, + { + "epoch": 1.3385532455476021, + "grad_norm": 0.0010225878795608878, + "learning_rate": 0.06770566618303668, + "loss": 0.0682, + "num_input_tokens_seen": 35517184, + "step": 27395 + }, + { + "epoch": 1.3387975472112965, + "grad_norm": 0.0011816531186923385, + "learning_rate": 0.06765642417776736, + "loss": 0.0641, + "num_input_tokens_seen": 35523424, + "step": 27400 + }, + { + "epoch": 1.3387975472112965, + "eval_loss": 0.08994396775960922, + "eval_runtime": 402.8505, + "eval_samples_per_second": 90.319, + "eval_steps_per_second": 22.582, + "num_input_tokens_seen": 35523424, + "step": 27400 + }, + { + "epoch": 1.339041848874991, + "grad_norm": 0.000678644108120352, + "learning_rate": 0.0676071948709122, + "loss": 0.0614, + "num_input_tokens_seen": 35529824, + "step": 27405 + }, + { + "epoch": 1.339286150538685, + "grad_norm": 0.0011382590746507049, + "learning_rate": 0.06755797827006307, + "loss": 0.0796, + "num_input_tokens_seen": 35536224, + "step": 27410 + }, + { + "epoch": 1.3395304522023794, + "grad_norm": 0.0010075075551867485, + "learning_rate": 0.06750877438280974, + "loss": 0.0778, + "num_input_tokens_seen": 35542368, + "step": 27415 + }, + { + "epoch": 1.3397747538660738, + "grad_norm": 0.0011160930152982473, + "learning_rate": 0.06745958321673998, + "loss": 0.0902, + "num_input_tokens_seen": 35548672, + "step": 27420 + }, + { + "epoch": 1.3400190555297682, + "grad_norm": 0.000993897789157927, + "learning_rate": 0.0674104047794398, + "loss": 0.0907, + "num_input_tokens_seen": 35554976, + "step": 27425 + }, + { + "epoch": 1.3402633571934626, + "grad_norm": 0.0014292413834482431, + "learning_rate": 0.06736123907849303, + "loss": 0.0868, + "num_input_tokens_seen": 35561088, + "step": 27430 + }, + { + "epoch": 1.340507658857157, + "grad_norm": 0.0017800374189391732, + "learning_rate": 0.06731208612148178, + "loss": 0.0671, + "num_input_tokens_seen": 35567680, + "step": 27435 + }, + { + "epoch": 1.340751960520851, + "grad_norm": 0.0025746794417500496, + "learning_rate": 0.0672629459159859, + "loss": 0.0951, + "num_input_tokens_seen": 35574016, + "step": 27440 + }, + { + "epoch": 1.3409962621845455, + "grad_norm": 0.001256619463674724, + "learning_rate": 0.0672138184695835, + "loss": 0.0764, + "num_input_tokens_seen": 35580608, + "step": 27445 + }, + { + "epoch": 1.3412405638482399, + "grad_norm": 0.0004612924240063876, + "learning_rate": 0.0671647037898507, + "loss": 0.0622, + "num_input_tokens_seen": 35587552, + "step": 27450 + }, + { + "epoch": 1.341484865511934, + "grad_norm": 0.0019710329361259937, + "learning_rate": 0.0671156018843615, + "loss": 0.0905, + "num_input_tokens_seen": 35593888, + "step": 27455 + }, + { + "epoch": 1.3417291671756284, + "grad_norm": 0.0007370646926574409, + "learning_rate": 0.06706651276068812, + "loss": 0.097, + "num_input_tokens_seen": 35600096, + "step": 27460 + }, + { + "epoch": 1.3419734688393228, + "grad_norm": 0.0021059869322925806, + "learning_rate": 0.06701743642640064, + "loss": 0.1008, + "num_input_tokens_seen": 35606080, + "step": 27465 + }, + { + "epoch": 1.3422177705030172, + "grad_norm": 0.0010928207775577903, + "learning_rate": 0.06696837288906729, + "loss": 0.0901, + "num_input_tokens_seen": 35612960, + "step": 27470 + }, + { + "epoch": 1.3424620721667115, + "grad_norm": 0.0022305576130747795, + "learning_rate": 0.06691932215625432, + "loss": 0.0825, + "num_input_tokens_seen": 35619136, + "step": 27475 + }, + { + "epoch": 1.342706373830406, + "grad_norm": 0.0014515273505821824, + "learning_rate": 0.06687028423552589, + "loss": 0.0708, + "num_input_tokens_seen": 35625728, + "step": 27480 + }, + { + "epoch": 1.3429506754941, + "grad_norm": 0.001581989461556077, + "learning_rate": 0.06682125913444435, + "loss": 0.1171, + "num_input_tokens_seen": 35632160, + "step": 27485 + }, + { + "epoch": 1.3431949771577945, + "grad_norm": 0.001185808563604951, + "learning_rate": 0.0667722468605699, + "loss": 0.092, + "num_input_tokens_seen": 35638944, + "step": 27490 + }, + { + "epoch": 1.3434392788214888, + "grad_norm": 0.0006531134131364524, + "learning_rate": 0.06672324742146094, + "loss": 0.0891, + "num_input_tokens_seen": 35645408, + "step": 27495 + }, + { + "epoch": 1.343683580485183, + "grad_norm": 0.001662932219915092, + "learning_rate": 0.06667426082467373, + "loss": 0.0961, + "num_input_tokens_seen": 35651552, + "step": 27500 + }, + { + "epoch": 1.3439278821488774, + "grad_norm": 0.0012755407951772213, + "learning_rate": 0.0666252870777626, + "loss": 0.1032, + "num_input_tokens_seen": 35658976, + "step": 27505 + }, + { + "epoch": 1.3441721838125718, + "grad_norm": 0.0013267322210595012, + "learning_rate": 0.06657632618827995, + "loss": 0.1101, + "num_input_tokens_seen": 35665280, + "step": 27510 + }, + { + "epoch": 1.3444164854762661, + "grad_norm": 0.0008687151712365448, + "learning_rate": 0.06652737816377623, + "loss": 0.0783, + "num_input_tokens_seen": 35671904, + "step": 27515 + }, + { + "epoch": 1.3446607871399605, + "grad_norm": 0.0014776919269934297, + "learning_rate": 0.06647844301179971, + "loss": 0.0871, + "num_input_tokens_seen": 35677888, + "step": 27520 + }, + { + "epoch": 1.344905088803655, + "grad_norm": 0.0012136960867792368, + "learning_rate": 0.06642952073989689, + "loss": 0.084, + "num_input_tokens_seen": 35684256, + "step": 27525 + }, + { + "epoch": 1.345149390467349, + "grad_norm": 0.0009473502286709845, + "learning_rate": 0.06638061135561223, + "loss": 0.0917, + "num_input_tokens_seen": 35690528, + "step": 27530 + }, + { + "epoch": 1.3453936921310434, + "grad_norm": 0.001632955390959978, + "learning_rate": 0.06633171486648808, + "loss": 0.0674, + "num_input_tokens_seen": 35697056, + "step": 27535 + }, + { + "epoch": 1.3456379937947378, + "grad_norm": 0.0015495619736611843, + "learning_rate": 0.06628283128006499, + "loss": 0.0809, + "num_input_tokens_seen": 35703552, + "step": 27540 + }, + { + "epoch": 1.345882295458432, + "grad_norm": 0.0010266402969136834, + "learning_rate": 0.0662339606038813, + "loss": 0.0716, + "num_input_tokens_seen": 35709760, + "step": 27545 + }, + { + "epoch": 1.3461265971221263, + "grad_norm": 0.000642311351839453, + "learning_rate": 0.06618510284547358, + "loss": 0.074, + "num_input_tokens_seen": 35715872, + "step": 27550 + }, + { + "epoch": 1.3463708987858207, + "grad_norm": 0.001322527532465756, + "learning_rate": 0.06613625801237633, + "loss": 0.0843, + "num_input_tokens_seen": 35722336, + "step": 27555 + }, + { + "epoch": 1.346615200449515, + "grad_norm": 0.0022053837310522795, + "learning_rate": 0.066087426112122, + "loss": 0.1009, + "num_input_tokens_seen": 35728704, + "step": 27560 + }, + { + "epoch": 1.3468595021132095, + "grad_norm": 0.0012695938348770142, + "learning_rate": 0.06603860715224101, + "loss": 0.0905, + "num_input_tokens_seen": 35735008, + "step": 27565 + }, + { + "epoch": 1.3471038037769036, + "grad_norm": 0.0010616264771670103, + "learning_rate": 0.06598980114026198, + "loss": 0.0634, + "num_input_tokens_seen": 35741664, + "step": 27570 + }, + { + "epoch": 1.347348105440598, + "grad_norm": 0.001025938312523067, + "learning_rate": 0.06594100808371128, + "loss": 0.0747, + "num_input_tokens_seen": 35748000, + "step": 27575 + }, + { + "epoch": 1.3475924071042924, + "grad_norm": 0.0017657442949712276, + "learning_rate": 0.06589222799011357, + "loss": 0.0737, + "num_input_tokens_seen": 35754336, + "step": 27580 + }, + { + "epoch": 1.3478367087679868, + "grad_norm": 0.0017273803241550922, + "learning_rate": 0.0658434608669912, + "loss": 0.0781, + "num_input_tokens_seen": 35760992, + "step": 27585 + }, + { + "epoch": 1.348081010431681, + "grad_norm": 0.002391078043729067, + "learning_rate": 0.06579470672186473, + "loss": 0.1042, + "num_input_tokens_seen": 35767872, + "step": 27590 + }, + { + "epoch": 1.3483253120953753, + "grad_norm": 0.0011382128577679396, + "learning_rate": 0.06574596556225275, + "loss": 0.0958, + "num_input_tokens_seen": 35774208, + "step": 27595 + }, + { + "epoch": 1.3485696137590697, + "grad_norm": 0.0010407684603706002, + "learning_rate": 0.06569723739567161, + "loss": 0.0644, + "num_input_tokens_seen": 35781024, + "step": 27600 + }, + { + "epoch": 1.3485696137590697, + "eval_loss": 0.09024789929389954, + "eval_runtime": 402.5787, + "eval_samples_per_second": 90.38, + "eval_steps_per_second": 22.597, + "num_input_tokens_seen": 35781024, + "step": 27600 + }, + { + "epoch": 1.348813915422764, + "grad_norm": 0.002329568611457944, + "learning_rate": 0.06564852222963588, + "loss": 0.0664, + "num_input_tokens_seen": 35787616, + "step": 27605 + }, + { + "epoch": 1.3490582170864585, + "grad_norm": 0.0013165851123631, + "learning_rate": 0.06559982007165813, + "loss": 0.0775, + "num_input_tokens_seen": 35793920, + "step": 27610 + }, + { + "epoch": 1.3493025187501526, + "grad_norm": 0.0013996383640915155, + "learning_rate": 0.06555113092924868, + "loss": 0.0844, + "num_input_tokens_seen": 35799936, + "step": 27615 + }, + { + "epoch": 1.349546820413847, + "grad_norm": 0.0012763350969180465, + "learning_rate": 0.06550245480991615, + "loss": 0.0964, + "num_input_tokens_seen": 35806560, + "step": 27620 + }, + { + "epoch": 1.3497911220775414, + "grad_norm": 0.001356414519250393, + "learning_rate": 0.0654537917211669, + "loss": 0.081, + "num_input_tokens_seen": 35813216, + "step": 27625 + }, + { + "epoch": 1.3500354237412358, + "grad_norm": 0.0006276586791500449, + "learning_rate": 0.0654051416705055, + "loss": 0.09, + "num_input_tokens_seen": 35819936, + "step": 27630 + }, + { + "epoch": 1.35027972540493, + "grad_norm": 0.0015570600517094135, + "learning_rate": 0.06535650466543427, + "loss": 0.108, + "num_input_tokens_seen": 35826208, + "step": 27635 + }, + { + "epoch": 1.3505240270686243, + "grad_norm": 0.001158856670372188, + "learning_rate": 0.0653078807134538, + "loss": 0.072, + "num_input_tokens_seen": 35833344, + "step": 27640 + }, + { + "epoch": 1.3507683287323187, + "grad_norm": 0.001970431534573436, + "learning_rate": 0.06525926982206236, + "loss": 0.0803, + "num_input_tokens_seen": 35839776, + "step": 27645 + }, + { + "epoch": 1.351012630396013, + "grad_norm": 0.0015311883762478828, + "learning_rate": 0.06521067199875648, + "loss": 0.0805, + "num_input_tokens_seen": 35846176, + "step": 27650 + }, + { + "epoch": 1.3512569320597074, + "grad_norm": 0.0012194332666695118, + "learning_rate": 0.06516208725103047, + "loss": 0.0942, + "num_input_tokens_seen": 35852576, + "step": 27655 + }, + { + "epoch": 1.3515012337234016, + "grad_norm": 0.001151514588855207, + "learning_rate": 0.06511351558637678, + "loss": 0.0892, + "num_input_tokens_seen": 35859072, + "step": 27660 + }, + { + "epoch": 1.351745535387096, + "grad_norm": 0.0013100354699417949, + "learning_rate": 0.06506495701228569, + "loss": 0.1111, + "num_input_tokens_seen": 35865184, + "step": 27665 + }, + { + "epoch": 1.3519898370507903, + "grad_norm": 0.0011344790691509843, + "learning_rate": 0.06501641153624559, + "loss": 0.1075, + "num_input_tokens_seen": 35871744, + "step": 27670 + }, + { + "epoch": 1.3522341387144847, + "grad_norm": 0.0015294412150979042, + "learning_rate": 0.06496787916574286, + "loss": 0.0824, + "num_input_tokens_seen": 35878208, + "step": 27675 + }, + { + "epoch": 1.3524784403781789, + "grad_norm": 0.000804751820396632, + "learning_rate": 0.06491935990826168, + "loss": 0.0757, + "num_input_tokens_seen": 35884512, + "step": 27680 + }, + { + "epoch": 1.3527227420418733, + "grad_norm": 0.0010659927502274513, + "learning_rate": 0.0648708537712844, + "loss": 0.0948, + "num_input_tokens_seen": 35890944, + "step": 27685 + }, + { + "epoch": 1.3529670437055676, + "grad_norm": 0.0014038265217095613, + "learning_rate": 0.06482236076229132, + "loss": 0.0998, + "num_input_tokens_seen": 35897472, + "step": 27690 + }, + { + "epoch": 1.353211345369262, + "grad_norm": 0.0007586674182675779, + "learning_rate": 0.06477388088876056, + "loss": 0.0746, + "num_input_tokens_seen": 35904320, + "step": 27695 + }, + { + "epoch": 1.3534556470329564, + "grad_norm": 0.0011202871100977063, + "learning_rate": 0.06472541415816846, + "loss": 0.0898, + "num_input_tokens_seen": 35910656, + "step": 27700 + }, + { + "epoch": 1.3536999486966506, + "grad_norm": 0.0014828471466898918, + "learning_rate": 0.06467696057798909, + "loss": 0.0631, + "num_input_tokens_seen": 35916896, + "step": 27705 + }, + { + "epoch": 1.353944250360345, + "grad_norm": 0.0015774646308273077, + "learning_rate": 0.0646285201556946, + "loss": 0.096, + "num_input_tokens_seen": 35923424, + "step": 27710 + }, + { + "epoch": 1.3541885520240393, + "grad_norm": 0.0014166977489367127, + "learning_rate": 0.06458009289875521, + "loss": 0.0783, + "num_input_tokens_seen": 35930368, + "step": 27715 + }, + { + "epoch": 1.3544328536877337, + "grad_norm": 0.0016802659956738353, + "learning_rate": 0.0645316788146389, + "loss": 0.0998, + "num_input_tokens_seen": 35936576, + "step": 27720 + }, + { + "epoch": 1.3546771553514279, + "grad_norm": 0.0006753075285814703, + "learning_rate": 0.06448327791081175, + "loss": 0.0686, + "num_input_tokens_seen": 35943072, + "step": 27725 + }, + { + "epoch": 1.3549214570151222, + "grad_norm": 0.0014536614762619138, + "learning_rate": 0.0644348901947379, + "loss": 0.1339, + "num_input_tokens_seen": 35948992, + "step": 27730 + }, + { + "epoch": 1.3551657586788166, + "grad_norm": 0.0016672369092702866, + "learning_rate": 0.06438651567387917, + "loss": 0.0802, + "num_input_tokens_seen": 35955296, + "step": 27735 + }, + { + "epoch": 1.355410060342511, + "grad_norm": 0.0012849883642047644, + "learning_rate": 0.0643381543556957, + "loss": 0.1083, + "num_input_tokens_seen": 35961536, + "step": 27740 + }, + { + "epoch": 1.3556543620062054, + "grad_norm": 0.0015659097116440535, + "learning_rate": 0.06428980624764526, + "loss": 0.1231, + "num_input_tokens_seen": 35967680, + "step": 27745 + }, + { + "epoch": 1.3558986636698995, + "grad_norm": 0.0017313881544396281, + "learning_rate": 0.06424147135718378, + "loss": 0.0906, + "num_input_tokens_seen": 35974528, + "step": 27750 + }, + { + "epoch": 1.356142965333594, + "grad_norm": 0.0016528692794963717, + "learning_rate": 0.06419314969176519, + "loss": 0.0778, + "num_input_tokens_seen": 35980960, + "step": 27755 + }, + { + "epoch": 1.3563872669972883, + "grad_norm": 0.001522575505077839, + "learning_rate": 0.06414484125884118, + "loss": 0.1286, + "num_input_tokens_seen": 35987136, + "step": 27760 + }, + { + "epoch": 1.3566315686609824, + "grad_norm": 0.0015727492282167077, + "learning_rate": 0.06409654606586157, + "loss": 0.0985, + "num_input_tokens_seen": 35993664, + "step": 27765 + }, + { + "epoch": 1.3568758703246768, + "grad_norm": 0.0007091779843904078, + "learning_rate": 0.06404826412027415, + "loss": 0.0869, + "num_input_tokens_seen": 36000384, + "step": 27770 + }, + { + "epoch": 1.3571201719883712, + "grad_norm": 0.0010004782816395164, + "learning_rate": 0.06399999542952453, + "loss": 0.0718, + "num_input_tokens_seen": 36006944, + "step": 27775 + }, + { + "epoch": 1.3573644736520656, + "grad_norm": 0.0009023742168210447, + "learning_rate": 0.0639517400010563, + "loss": 0.073, + "num_input_tokens_seen": 36013600, + "step": 27780 + }, + { + "epoch": 1.35760877531576, + "grad_norm": 0.0008886153809726238, + "learning_rate": 0.06390349784231118, + "loss": 0.0691, + "num_input_tokens_seen": 36020544, + "step": 27785 + }, + { + "epoch": 1.3578530769794543, + "grad_norm": 0.0007031045388430357, + "learning_rate": 0.06385526896072859, + "loss": 0.0723, + "num_input_tokens_seen": 36027264, + "step": 27790 + }, + { + "epoch": 1.3580973786431485, + "grad_norm": 0.00047846572124399245, + "learning_rate": 0.06380705336374613, + "loss": 0.0749, + "num_input_tokens_seen": 36033696, + "step": 27795 + }, + { + "epoch": 1.3583416803068429, + "grad_norm": 0.0009639273630455136, + "learning_rate": 0.06375885105879918, + "loss": 0.0943, + "num_input_tokens_seen": 36040224, + "step": 27800 + }, + { + "epoch": 1.3583416803068429, + "eval_loss": 0.08922871947288513, + "eval_runtime": 401.8257, + "eval_samples_per_second": 90.549, + "eval_steps_per_second": 22.639, + "num_input_tokens_seen": 36040224, + "step": 27800 + }, + { + "epoch": 1.3585859819705373, + "grad_norm": 0.0012717739446088672, + "learning_rate": 0.06371066205332115, + "loss": 0.0921, + "num_input_tokens_seen": 36046688, + "step": 27805 + }, + { + "epoch": 1.3588302836342314, + "grad_norm": 0.001397769316099584, + "learning_rate": 0.06366248635474347, + "loss": 0.0966, + "num_input_tokens_seen": 36052928, + "step": 27810 + }, + { + "epoch": 1.3590745852979258, + "grad_norm": 0.0013784259790554643, + "learning_rate": 0.06361432397049532, + "loss": 0.1168, + "num_input_tokens_seen": 36059552, + "step": 27815 + }, + { + "epoch": 1.3593188869616202, + "grad_norm": 0.0010807664366438985, + "learning_rate": 0.06356617490800408, + "loss": 0.0738, + "num_input_tokens_seen": 36066368, + "step": 27820 + }, + { + "epoch": 1.3595631886253146, + "grad_norm": 0.0007257935940288007, + "learning_rate": 0.06351803917469478, + "loss": 0.0847, + "num_input_tokens_seen": 36073088, + "step": 27825 + }, + { + "epoch": 1.359807490289009, + "grad_norm": 0.0019748082850128412, + "learning_rate": 0.06346991677799067, + "loss": 0.0715, + "num_input_tokens_seen": 36079520, + "step": 27830 + }, + { + "epoch": 1.3600517919527033, + "grad_norm": 0.0011830959701910615, + "learning_rate": 0.06342180772531283, + "loss": 0.0677, + "num_input_tokens_seen": 36086464, + "step": 27835 + }, + { + "epoch": 1.3602960936163975, + "grad_norm": 0.0010325255570933223, + "learning_rate": 0.06337371202408021, + "loss": 0.0681, + "num_input_tokens_seen": 36092768, + "step": 27840 + }, + { + "epoch": 1.3605403952800919, + "grad_norm": 0.0007732262019999325, + "learning_rate": 0.06332562968170984, + "loss": 0.0804, + "num_input_tokens_seen": 36098912, + "step": 27845 + }, + { + "epoch": 1.3607846969437862, + "grad_norm": 0.0014943901915103197, + "learning_rate": 0.06327756070561656, + "loss": 0.0961, + "num_input_tokens_seen": 36105696, + "step": 27850 + }, + { + "epoch": 1.3610289986074804, + "grad_norm": 0.0015549794770777225, + "learning_rate": 0.06322950510321329, + "loss": 0.0949, + "num_input_tokens_seen": 36112064, + "step": 27855 + }, + { + "epoch": 1.3612733002711748, + "grad_norm": 0.0012212564470246434, + "learning_rate": 0.06318146288191076, + "loss": 0.0833, + "num_input_tokens_seen": 36119072, + "step": 27860 + }, + { + "epoch": 1.3615176019348691, + "grad_norm": 0.0006683591054752469, + "learning_rate": 0.06313343404911763, + "loss": 0.0961, + "num_input_tokens_seen": 36125440, + "step": 27865 + }, + { + "epoch": 1.3617619035985635, + "grad_norm": 0.0009955615969374776, + "learning_rate": 0.0630854186122406, + "loss": 0.0895, + "num_input_tokens_seen": 36131840, + "step": 27870 + }, + { + "epoch": 1.362006205262258, + "grad_norm": 0.0020263190381228924, + "learning_rate": 0.06303741657868431, + "loss": 0.0981, + "num_input_tokens_seen": 36138144, + "step": 27875 + }, + { + "epoch": 1.3622505069259523, + "grad_norm": 0.0021800242830067873, + "learning_rate": 0.06298942795585115, + "loss": 0.1111, + "num_input_tokens_seen": 36144448, + "step": 27880 + }, + { + "epoch": 1.3624948085896464, + "grad_norm": 0.001056336099281907, + "learning_rate": 0.06294145275114167, + "loss": 0.0859, + "num_input_tokens_seen": 36150688, + "step": 27885 + }, + { + "epoch": 1.3627391102533408, + "grad_norm": 0.001235118368640542, + "learning_rate": 0.06289349097195428, + "loss": 0.0724, + "num_input_tokens_seen": 36156928, + "step": 27890 + }, + { + "epoch": 1.3629834119170352, + "grad_norm": 0.00173288700170815, + "learning_rate": 0.06284554262568516, + "loss": 0.1088, + "num_input_tokens_seen": 36163136, + "step": 27895 + }, + { + "epoch": 1.3632277135807294, + "grad_norm": 0.0018336707726120949, + "learning_rate": 0.06279760771972868, + "loss": 0.0818, + "num_input_tokens_seen": 36169376, + "step": 27900 + }, + { + "epoch": 1.3634720152444237, + "grad_norm": 0.0017393151065334678, + "learning_rate": 0.06274968626147688, + "loss": 0.1055, + "num_input_tokens_seen": 36175744, + "step": 27905 + }, + { + "epoch": 1.3637163169081181, + "grad_norm": 0.001563719124533236, + "learning_rate": 0.06270177825831993, + "loss": 0.0697, + "num_input_tokens_seen": 36182560, + "step": 27910 + }, + { + "epoch": 1.3639606185718125, + "grad_norm": 0.0019385053310543299, + "learning_rate": 0.06265388371764587, + "loss": 0.0893, + "num_input_tokens_seen": 36188992, + "step": 27915 + }, + { + "epoch": 1.3642049202355069, + "grad_norm": 0.0009982433402910829, + "learning_rate": 0.0626060026468406, + "loss": 0.0759, + "num_input_tokens_seen": 36195136, + "step": 27920 + }, + { + "epoch": 1.3644492218992013, + "grad_norm": 0.0011466332944110036, + "learning_rate": 0.06255813505328794, + "loss": 0.0883, + "num_input_tokens_seen": 36201664, + "step": 27925 + }, + { + "epoch": 1.3646935235628954, + "grad_norm": 0.0006821415154263377, + "learning_rate": 0.06251028094436978, + "loss": 0.0769, + "num_input_tokens_seen": 36208128, + "step": 27930 + }, + { + "epoch": 1.3649378252265898, + "grad_norm": 0.0016194566851481795, + "learning_rate": 0.06246244032746568, + "loss": 0.0852, + "num_input_tokens_seen": 36214304, + "step": 27935 + }, + { + "epoch": 1.3651821268902842, + "grad_norm": 0.0014966689050197601, + "learning_rate": 0.06241461320995342, + "loss": 0.0943, + "num_input_tokens_seen": 36221056, + "step": 27940 + }, + { + "epoch": 1.3654264285539783, + "grad_norm": 0.0011382513912394643, + "learning_rate": 0.062366799599208426, + "loss": 0.0856, + "num_input_tokens_seen": 36227584, + "step": 27945 + }, + { + "epoch": 1.3656707302176727, + "grad_norm": 0.0014910238096490502, + "learning_rate": 0.06231899950260418, + "loss": 0.0926, + "num_input_tokens_seen": 36233760, + "step": 27950 + }, + { + "epoch": 1.365915031881367, + "grad_norm": 0.0008412337629124522, + "learning_rate": 0.06227121292751214, + "loss": 0.0578, + "num_input_tokens_seen": 36239872, + "step": 27955 + }, + { + "epoch": 1.3661593335450615, + "grad_norm": 0.0009373155771754682, + "learning_rate": 0.062223439881301496, + "loss": 0.0634, + "num_input_tokens_seen": 36246144, + "step": 27960 + }, + { + "epoch": 1.3664036352087559, + "grad_norm": 0.0012367871822789311, + "learning_rate": 0.06217568037133948, + "loss": 0.0989, + "num_input_tokens_seen": 36252928, + "step": 27965 + }, + { + "epoch": 1.3666479368724502, + "grad_norm": 0.0011689055245369673, + "learning_rate": 0.06212793440499126, + "loss": 0.0936, + "num_input_tokens_seen": 36259264, + "step": 27970 + }, + { + "epoch": 1.3668922385361444, + "grad_norm": 0.0012893362436443567, + "learning_rate": 0.062080201989619783, + "loss": 0.0905, + "num_input_tokens_seen": 36265856, + "step": 27975 + }, + { + "epoch": 1.3671365401998388, + "grad_norm": 0.0010894249426200986, + "learning_rate": 0.062032483132586094, + "loss": 0.0932, + "num_input_tokens_seen": 36272224, + "step": 27980 + }, + { + "epoch": 1.3673808418635331, + "grad_norm": 0.00041186410817317665, + "learning_rate": 0.0619847778412489, + "loss": 0.1108, + "num_input_tokens_seen": 36278336, + "step": 27985 + }, + { + "epoch": 1.3676251435272273, + "grad_norm": 0.0011094348737969995, + "learning_rate": 0.06193708612296509, + "loss": 0.0799, + "num_input_tokens_seen": 36284704, + "step": 27990 + }, + { + "epoch": 1.3678694451909217, + "grad_norm": 0.001080686692148447, + "learning_rate": 0.06188940798508923, + "loss": 0.0705, + "num_input_tokens_seen": 36290784, + "step": 27995 + }, + { + "epoch": 1.368113746854616, + "grad_norm": 0.0008682747138664126, + "learning_rate": 0.06184174343497397, + "loss": 0.066, + "num_input_tokens_seen": 36297952, + "step": 28000 + }, + { + "epoch": 1.368113746854616, + "eval_loss": 0.09009172022342682, + "eval_runtime": 402.0356, + "eval_samples_per_second": 90.502, + "eval_steps_per_second": 22.627, + "num_input_tokens_seen": 36297952, + "step": 28000 + }, + { + "epoch": 1.3683580485183104, + "grad_norm": 0.0006706152344122529, + "learning_rate": 0.061794092479969726, + "loss": 0.0604, + "num_input_tokens_seen": 36304512, + "step": 28005 + }, + { + "epoch": 1.3686023501820048, + "grad_norm": 0.001285684877075255, + "learning_rate": 0.06174645512742485, + "loss": 0.0763, + "num_input_tokens_seen": 36310432, + "step": 28010 + }, + { + "epoch": 1.3688466518456992, + "grad_norm": 0.0014535292284563184, + "learning_rate": 0.06169883138468565, + "loss": 0.0868, + "num_input_tokens_seen": 36317376, + "step": 28015 + }, + { + "epoch": 1.3690909535093934, + "grad_norm": 0.0011757098836824298, + "learning_rate": 0.06165122125909637, + "loss": 0.1065, + "num_input_tokens_seen": 36323968, + "step": 28020 + }, + { + "epoch": 1.3693352551730877, + "grad_norm": 0.0013645458966493607, + "learning_rate": 0.061603624757998965, + "loss": 0.0852, + "num_input_tokens_seen": 36330240, + "step": 28025 + }, + { + "epoch": 1.3695795568367821, + "grad_norm": 0.0007354200934059918, + "learning_rate": 0.0615560418887335, + "loss": 0.0656, + "num_input_tokens_seen": 36336704, + "step": 28030 + }, + { + "epoch": 1.3698238585004763, + "grad_norm": 0.0014475708594545722, + "learning_rate": 0.06150847265863787, + "loss": 0.1092, + "num_input_tokens_seen": 36343296, + "step": 28035 + }, + { + "epoch": 1.3700681601641707, + "grad_norm": 0.0010707227047532797, + "learning_rate": 0.061460917075047757, + "loss": 0.0775, + "num_input_tokens_seen": 36349504, + "step": 28040 + }, + { + "epoch": 1.370312461827865, + "grad_norm": 0.000860165455378592, + "learning_rate": 0.06141337514529694, + "loss": 0.0646, + "num_input_tokens_seen": 36356128, + "step": 28045 + }, + { + "epoch": 1.3705567634915594, + "grad_norm": 0.00100353779271245, + "learning_rate": 0.06136584687671687, + "loss": 0.0815, + "num_input_tokens_seen": 36362336, + "step": 28050 + }, + { + "epoch": 1.3708010651552538, + "grad_norm": 0.001576851005665958, + "learning_rate": 0.061318332276637064, + "loss": 0.087, + "num_input_tokens_seen": 36368512, + "step": 28055 + }, + { + "epoch": 1.3710453668189482, + "grad_norm": 0.003078057197853923, + "learning_rate": 0.06127083135238491, + "loss": 0.0926, + "num_input_tokens_seen": 36374656, + "step": 28060 + }, + { + "epoch": 1.3712896684826423, + "grad_norm": 0.0013888776302337646, + "learning_rate": 0.06122334411128555, + "loss": 0.0768, + "num_input_tokens_seen": 36380928, + "step": 28065 + }, + { + "epoch": 1.3715339701463367, + "grad_norm": 0.001901148003526032, + "learning_rate": 0.06117587056066223, + "loss": 0.0762, + "num_input_tokens_seen": 36388000, + "step": 28070 + }, + { + "epoch": 1.371778271810031, + "grad_norm": 0.0008613595273345709, + "learning_rate": 0.06112841070783589, + "loss": 0.0735, + "num_input_tokens_seen": 36394368, + "step": 28075 + }, + { + "epoch": 1.3720225734737252, + "grad_norm": 0.0007747175404801965, + "learning_rate": 0.061080964560125406, + "loss": 0.0854, + "num_input_tokens_seen": 36400704, + "step": 28080 + }, + { + "epoch": 1.3722668751374196, + "grad_norm": 0.0015227393014356494, + "learning_rate": 0.06103353212484766, + "loss": 0.0933, + "num_input_tokens_seen": 36407104, + "step": 28085 + }, + { + "epoch": 1.372511176801114, + "grad_norm": 0.0016606012359261513, + "learning_rate": 0.06098611340931722, + "loss": 0.0947, + "num_input_tokens_seen": 36413600, + "step": 28090 + }, + { + "epoch": 1.3727554784648084, + "grad_norm": 0.0011238135630264878, + "learning_rate": 0.06093870842084672, + "loss": 0.0972, + "num_input_tokens_seen": 36419968, + "step": 28095 + }, + { + "epoch": 1.3729997801285028, + "grad_norm": 0.000994914909824729, + "learning_rate": 0.06089131716674666, + "loss": 0.0678, + "num_input_tokens_seen": 36426624, + "step": 28100 + }, + { + "epoch": 1.373244081792197, + "grad_norm": 0.0015723173273727298, + "learning_rate": 0.060843939654325226, + "loss": 0.0749, + "num_input_tokens_seen": 36433184, + "step": 28105 + }, + { + "epoch": 1.3734883834558913, + "grad_norm": 0.001226854044944048, + "learning_rate": 0.06079657589088873, + "loss": 0.0982, + "num_input_tokens_seen": 36439744, + "step": 28110 + }, + { + "epoch": 1.3737326851195857, + "grad_norm": 0.0009220019564963877, + "learning_rate": 0.06074922588374126, + "loss": 0.0838, + "num_input_tokens_seen": 36446240, + "step": 28115 + }, + { + "epoch": 1.37397698678328, + "grad_norm": 0.0016804899787530303, + "learning_rate": 0.06070188964018472, + "loss": 0.0876, + "num_input_tokens_seen": 36452512, + "step": 28120 + }, + { + "epoch": 1.3742212884469742, + "grad_norm": 0.0016855106223374605, + "learning_rate": 0.06065456716751902, + "loss": 0.0748, + "num_input_tokens_seen": 36459360, + "step": 28125 + }, + { + "epoch": 1.3744655901106686, + "grad_norm": 0.0020971661433577538, + "learning_rate": 0.06060725847304182, + "loss": 0.0842, + "num_input_tokens_seen": 36465536, + "step": 28130 + }, + { + "epoch": 1.374709891774363, + "grad_norm": 0.0011097536189481616, + "learning_rate": 0.06055996356404877, + "loss": 0.0745, + "num_input_tokens_seen": 36472192, + "step": 28135 + }, + { + "epoch": 1.3749541934380574, + "grad_norm": 0.0007392533007077873, + "learning_rate": 0.06051268244783327, + "loss": 0.0892, + "num_input_tokens_seen": 36478816, + "step": 28140 + }, + { + "epoch": 1.3751984951017517, + "grad_norm": 0.0006934689008630812, + "learning_rate": 0.06046541513168676, + "loss": 0.0699, + "num_input_tokens_seen": 36485376, + "step": 28145 + }, + { + "epoch": 1.375442796765446, + "grad_norm": 0.0011629121145233512, + "learning_rate": 0.060418161622898356, + "loss": 0.0698, + "num_input_tokens_seen": 36491648, + "step": 28150 + }, + { + "epoch": 1.3756870984291403, + "grad_norm": 0.0012167161330580711, + "learning_rate": 0.06037092192875521, + "loss": 0.0679, + "num_input_tokens_seen": 36498688, + "step": 28155 + }, + { + "epoch": 1.3759314000928347, + "grad_norm": 0.0010463251965120435, + "learning_rate": 0.060323696056542225, + "loss": 0.1058, + "num_input_tokens_seen": 36505312, + "step": 28160 + }, + { + "epoch": 1.376175701756529, + "grad_norm": 0.0019138803472742438, + "learning_rate": 0.06027648401354229, + "loss": 0.0757, + "num_input_tokens_seen": 36511360, + "step": 28165 + }, + { + "epoch": 1.3764200034202232, + "grad_norm": 0.001557568204589188, + "learning_rate": 0.06022928580703601, + "loss": 0.0877, + "num_input_tokens_seen": 36518752, + "step": 28170 + }, + { + "epoch": 1.3766643050839176, + "grad_norm": 0.0014171833172440529, + "learning_rate": 0.060182101444301986, + "loss": 0.0719, + "num_input_tokens_seen": 36524992, + "step": 28175 + }, + { + "epoch": 1.376908606747612, + "grad_norm": 0.0018651343416422606, + "learning_rate": 0.06013493093261669, + "loss": 0.0849, + "num_input_tokens_seen": 36531648, + "step": 28180 + }, + { + "epoch": 1.3771529084113063, + "grad_norm": 0.0010321710724383593, + "learning_rate": 0.06008777427925432, + "loss": 0.0973, + "num_input_tokens_seen": 36537856, + "step": 28185 + }, + { + "epoch": 1.3773972100750007, + "grad_norm": 0.0010068804258480668, + "learning_rate": 0.06004063149148705, + "loss": 0.0958, + "num_input_tokens_seen": 36544096, + "step": 28190 + }, + { + "epoch": 1.3776415117386949, + "grad_norm": 0.0010428217938169837, + "learning_rate": 0.05999350257658497, + "loss": 0.0693, + "num_input_tokens_seen": 36550688, + "step": 28195 + }, + { + "epoch": 1.3778858134023892, + "grad_norm": 0.0015801152912899852, + "learning_rate": 0.05994638754181582, + "loss": 0.1466, + "num_input_tokens_seen": 36557056, + "step": 28200 + }, + { + "epoch": 1.3778858134023892, + "eval_loss": 0.08804360777139664, + "eval_runtime": 402.3203, + "eval_samples_per_second": 90.438, + "eval_steps_per_second": 22.611, + "num_input_tokens_seen": 36557056, + "step": 28200 + }, + { + "epoch": 1.3781301150660836, + "grad_norm": 0.001594768138602376, + "learning_rate": 0.059899286394445445, + "loss": 0.0926, + "num_input_tokens_seen": 36563616, + "step": 28205 + }, + { + "epoch": 1.378374416729778, + "grad_norm": 0.0014098254032433033, + "learning_rate": 0.059852199141737346, + "loss": 0.0679, + "num_input_tokens_seen": 36570048, + "step": 28210 + }, + { + "epoch": 1.3786187183934722, + "grad_norm": 0.0009340434335172176, + "learning_rate": 0.05980512579095304, + "loss": 0.0958, + "num_input_tokens_seen": 36576128, + "step": 28215 + }, + { + "epoch": 1.3788630200571665, + "grad_norm": 0.0011819856008514762, + "learning_rate": 0.05975806634935181, + "loss": 0.1093, + "num_input_tokens_seen": 36582720, + "step": 28220 + }, + { + "epoch": 1.379107321720861, + "grad_norm": 0.00082395103527233, + "learning_rate": 0.05971102082419076, + "loss": 0.0848, + "num_input_tokens_seen": 36588832, + "step": 28225 + }, + { + "epoch": 1.3793516233845553, + "grad_norm": 0.0013325877953320742, + "learning_rate": 0.05966398922272492, + "loss": 0.0983, + "num_input_tokens_seen": 36595424, + "step": 28230 + }, + { + "epoch": 1.3795959250482497, + "grad_norm": 0.0014879563823342323, + "learning_rate": 0.059616971552207236, + "loss": 0.0791, + "num_input_tokens_seen": 36602144, + "step": 28235 + }, + { + "epoch": 1.3798402267119438, + "grad_norm": 0.00104216777253896, + "learning_rate": 0.059569967819888305, + "loss": 0.0615, + "num_input_tokens_seen": 36608544, + "step": 28240 + }, + { + "epoch": 1.3800845283756382, + "grad_norm": 0.0012143630301579833, + "learning_rate": 0.05952297803301681, + "loss": 0.0728, + "num_input_tokens_seen": 36614944, + "step": 28245 + }, + { + "epoch": 1.3803288300393326, + "grad_norm": 0.00267794425599277, + "learning_rate": 0.059476002198839056, + "loss": 0.0864, + "num_input_tokens_seen": 36621440, + "step": 28250 + }, + { + "epoch": 1.380573131703027, + "grad_norm": 0.0011296082520857453, + "learning_rate": 0.05942904032459935, + "loss": 0.1339, + "num_input_tokens_seen": 36627744, + "step": 28255 + }, + { + "epoch": 1.3808174333667211, + "grad_norm": 0.0016766919288784266, + "learning_rate": 0.05938209241753987, + "loss": 0.089, + "num_input_tokens_seen": 36634240, + "step": 28260 + }, + { + "epoch": 1.3810617350304155, + "grad_norm": 0.0014619337162002921, + "learning_rate": 0.05933515848490046, + "loss": 0.078, + "num_input_tokens_seen": 36640992, + "step": 28265 + }, + { + "epoch": 1.38130603669411, + "grad_norm": 0.0018239483470097184, + "learning_rate": 0.059288238533918985, + "loss": 0.1041, + "num_input_tokens_seen": 36647296, + "step": 28270 + }, + { + "epoch": 1.3815503383578043, + "grad_norm": 0.0016804294427856803, + "learning_rate": 0.05924133257183113, + "loss": 0.0944, + "num_input_tokens_seen": 36653568, + "step": 28275 + }, + { + "epoch": 1.3817946400214987, + "grad_norm": 0.0013727350160479546, + "learning_rate": 0.059194440605870285, + "loss": 0.0711, + "num_input_tokens_seen": 36660352, + "step": 28280 + }, + { + "epoch": 1.3820389416851928, + "grad_norm": 0.0014828077983111143, + "learning_rate": 0.059147562643267884, + "loss": 0.08, + "num_input_tokens_seen": 36666912, + "step": 28285 + }, + { + "epoch": 1.3822832433488872, + "grad_norm": 0.001321942312642932, + "learning_rate": 0.059100698691253055, + "loss": 0.084, + "num_input_tokens_seen": 36673216, + "step": 28290 + }, + { + "epoch": 1.3825275450125816, + "grad_norm": 0.001986920600757003, + "learning_rate": 0.05905384875705273, + "loss": 0.0775, + "num_input_tokens_seen": 36679904, + "step": 28295 + }, + { + "epoch": 1.3827718466762757, + "grad_norm": 0.0015643128426745534, + "learning_rate": 0.05900701284789189, + "loss": 0.0903, + "num_input_tokens_seen": 36686112, + "step": 28300 + }, + { + "epoch": 1.38301614833997, + "grad_norm": 0.0013855474535375834, + "learning_rate": 0.058960190970993115, + "loss": 0.0805, + "num_input_tokens_seen": 36693056, + "step": 28305 + }, + { + "epoch": 1.3832604500036645, + "grad_norm": 0.0016649194294586778, + "learning_rate": 0.058913383133576955, + "loss": 0.0748, + "num_input_tokens_seen": 36699456, + "step": 28310 + }, + { + "epoch": 1.3835047516673589, + "grad_norm": 0.0018881605938076973, + "learning_rate": 0.05886658934286185, + "loss": 0.0891, + "num_input_tokens_seen": 36705920, + "step": 28315 + }, + { + "epoch": 1.3837490533310532, + "grad_norm": 0.0021274001337587833, + "learning_rate": 0.058819809606063846, + "loss": 0.1099, + "num_input_tokens_seen": 36712960, + "step": 28320 + }, + { + "epoch": 1.3839933549947476, + "grad_norm": 0.0012218469055369496, + "learning_rate": 0.05877304393039711, + "loss": 0.0665, + "num_input_tokens_seen": 36720064, + "step": 28325 + }, + { + "epoch": 1.3842376566584418, + "grad_norm": 0.0012214789167046547, + "learning_rate": 0.05872629232307338, + "loss": 0.0851, + "num_input_tokens_seen": 36726464, + "step": 28330 + }, + { + "epoch": 1.3844819583221362, + "grad_norm": 0.0007010240806266665, + "learning_rate": 0.05867955479130239, + "loss": 0.0836, + "num_input_tokens_seen": 36732640, + "step": 28335 + }, + { + "epoch": 1.3847262599858305, + "grad_norm": 0.000889610790181905, + "learning_rate": 0.058632831342291705, + "loss": 0.0731, + "num_input_tokens_seen": 36739200, + "step": 28340 + }, + { + "epoch": 1.3849705616495247, + "grad_norm": 0.0006328709423542023, + "learning_rate": 0.05858612198324655, + "loss": 0.0601, + "num_input_tokens_seen": 36746144, + "step": 28345 + }, + { + "epoch": 1.385214863313219, + "grad_norm": 0.0008902396075427532, + "learning_rate": 0.05853942672137025, + "loss": 0.0891, + "num_input_tokens_seen": 36752672, + "step": 28350 + }, + { + "epoch": 1.3854591649769135, + "grad_norm": 0.0011598184937611222, + "learning_rate": 0.05849274556386363, + "loss": 0.0815, + "num_input_tokens_seen": 36758560, + "step": 28355 + }, + { + "epoch": 1.3857034666406078, + "grad_norm": 0.0014235101407393813, + "learning_rate": 0.05844607851792567, + "loss": 0.0788, + "num_input_tokens_seen": 36765344, + "step": 28360 + }, + { + "epoch": 1.3859477683043022, + "grad_norm": 0.0016387980431318283, + "learning_rate": 0.058399425590752924, + "loss": 0.0855, + "num_input_tokens_seen": 36771456, + "step": 28365 + }, + { + "epoch": 1.3861920699679966, + "grad_norm": 0.0013268992770463228, + "learning_rate": 0.05835278678953985, + "loss": 0.0609, + "num_input_tokens_seen": 36777824, + "step": 28370 + }, + { + "epoch": 1.3864363716316908, + "grad_norm": 0.001323521020822227, + "learning_rate": 0.05830616212147874, + "loss": 0.0918, + "num_input_tokens_seen": 36784448, + "step": 28375 + }, + { + "epoch": 1.3866806732953851, + "grad_norm": 0.0017904428532347083, + "learning_rate": 0.058259551593759784, + "loss": 0.0693, + "num_input_tokens_seen": 36790720, + "step": 28380 + }, + { + "epoch": 1.3869249749590795, + "grad_norm": 0.0022853254340589046, + "learning_rate": 0.058212955213570804, + "loss": 0.1024, + "num_input_tokens_seen": 36797344, + "step": 28385 + }, + { + "epoch": 1.3871692766227737, + "grad_norm": 0.0017817504703998566, + "learning_rate": 0.0581663729880976, + "loss": 0.0929, + "num_input_tokens_seen": 36803424, + "step": 28390 + }, + { + "epoch": 1.387413578286468, + "grad_norm": 0.0014560221461579204, + "learning_rate": 0.05811980492452379, + "loss": 0.094, + "num_input_tokens_seen": 36809632, + "step": 28395 + }, + { + "epoch": 1.3876578799501624, + "grad_norm": 0.0009016605908982456, + "learning_rate": 0.058073251030030644, + "loss": 0.109, + "num_input_tokens_seen": 36815904, + "step": 28400 + }, + { + "epoch": 1.3876578799501624, + "eval_loss": 0.08811922371387482, + "eval_runtime": 402.386, + "eval_samples_per_second": 90.423, + "eval_steps_per_second": 22.608, + "num_input_tokens_seen": 36815904, + "step": 28400 + }, + { + "epoch": 1.3879021816138568, + "grad_norm": 0.0015694027533754706, + "learning_rate": 0.05802671131179747, + "loss": 0.0833, + "num_input_tokens_seen": 36822240, + "step": 28405 + }, + { + "epoch": 1.3881464832775512, + "grad_norm": 0.0009453463135287166, + "learning_rate": 0.057980185777001154, + "loss": 0.0795, + "num_input_tokens_seen": 36828480, + "step": 28410 + }, + { + "epoch": 1.3883907849412456, + "grad_norm": 0.00121788471005857, + "learning_rate": 0.057933674432816606, + "loss": 0.0973, + "num_input_tokens_seen": 36834944, + "step": 28415 + }, + { + "epoch": 1.3886350866049397, + "grad_norm": 0.0008735384326428175, + "learning_rate": 0.05788717728641648, + "loss": 0.0881, + "num_input_tokens_seen": 36841248, + "step": 28420 + }, + { + "epoch": 1.388879388268634, + "grad_norm": 0.0010026677045971155, + "learning_rate": 0.057840694344971126, + "loss": 0.0962, + "num_input_tokens_seen": 36847552, + "step": 28425 + }, + { + "epoch": 1.3891236899323285, + "grad_norm": 0.001822098158299923, + "learning_rate": 0.0577942256156489, + "loss": 0.0883, + "num_input_tokens_seen": 36854208, + "step": 28430 + }, + { + "epoch": 1.3893679915960226, + "grad_norm": 0.0011928032618016005, + "learning_rate": 0.057747771105615804, + "loss": 0.0863, + "num_input_tokens_seen": 36860320, + "step": 28435 + }, + { + "epoch": 1.389612293259717, + "grad_norm": 0.0013689729385077953, + "learning_rate": 0.05770133082203568, + "loss": 0.0886, + "num_input_tokens_seen": 36867136, + "step": 28440 + }, + { + "epoch": 1.3898565949234114, + "grad_norm": 0.0017490154132246971, + "learning_rate": 0.0576549047720703, + "loss": 0.1306, + "num_input_tokens_seen": 36873632, + "step": 28445 + }, + { + "epoch": 1.3901008965871058, + "grad_norm": 0.0007920098141767085, + "learning_rate": 0.05760849296287902, + "loss": 0.0703, + "num_input_tokens_seen": 36880352, + "step": 28450 + }, + { + "epoch": 1.3903451982508002, + "grad_norm": 0.0014837330672889948, + "learning_rate": 0.05756209540161919, + "loss": 0.109, + "num_input_tokens_seen": 36886784, + "step": 28455 + }, + { + "epoch": 1.3905894999144945, + "grad_norm": 0.0007720084977336228, + "learning_rate": 0.05751571209544595, + "loss": 0.0765, + "num_input_tokens_seen": 36893312, + "step": 28460 + }, + { + "epoch": 1.3908338015781887, + "grad_norm": 0.0008356455364264548, + "learning_rate": 0.057469343051512085, + "loss": 0.0543, + "num_input_tokens_seen": 36899840, + "step": 28465 + }, + { + "epoch": 1.391078103241883, + "grad_norm": 0.002460371470078826, + "learning_rate": 0.057422988276968324, + "loss": 0.1109, + "num_input_tokens_seen": 36906144, + "step": 28470 + }, + { + "epoch": 1.3913224049055775, + "grad_norm": 0.0017221012385562062, + "learning_rate": 0.05737664777896323, + "loss": 0.1159, + "num_input_tokens_seen": 36912448, + "step": 28475 + }, + { + "epoch": 1.3915667065692716, + "grad_norm": 0.0010390261886641383, + "learning_rate": 0.057330321564642975, + "loss": 0.0646, + "num_input_tokens_seen": 36919168, + "step": 28480 + }, + { + "epoch": 1.391811008232966, + "grad_norm": 0.0016106113325804472, + "learning_rate": 0.05728400964115174, + "loss": 0.1156, + "num_input_tokens_seen": 36925728, + "step": 28485 + }, + { + "epoch": 1.3920553098966604, + "grad_norm": 0.001967286691069603, + "learning_rate": 0.057237712015631305, + "loss": 0.1007, + "num_input_tokens_seen": 36932128, + "step": 28490 + }, + { + "epoch": 1.3922996115603548, + "grad_norm": 0.0013463431969285011, + "learning_rate": 0.057191428695221425, + "loss": 0.1234, + "num_input_tokens_seen": 36938432, + "step": 28495 + }, + { + "epoch": 1.3925439132240491, + "grad_norm": 0.0006450550281442702, + "learning_rate": 0.05714515968705958, + "loss": 0.0679, + "num_input_tokens_seen": 36945088, + "step": 28500 + }, + { + "epoch": 1.3927882148877435, + "grad_norm": 0.0010170727036893368, + "learning_rate": 0.05709890499828099, + "loss": 0.0794, + "num_input_tokens_seen": 36951584, + "step": 28505 + }, + { + "epoch": 1.3930325165514377, + "grad_norm": 0.0014699131716042757, + "learning_rate": 0.05705266463601868, + "loss": 0.0788, + "num_input_tokens_seen": 36957952, + "step": 28510 + }, + { + "epoch": 1.393276818215132, + "grad_norm": 0.001171488082036376, + "learning_rate": 0.057006438607403565, + "loss": 0.0978, + "num_input_tokens_seen": 36964576, + "step": 28515 + }, + { + "epoch": 1.3935211198788264, + "grad_norm": 0.0016148254508152604, + "learning_rate": 0.056960226919564205, + "loss": 0.0949, + "num_input_tokens_seen": 36971104, + "step": 28520 + }, + { + "epoch": 1.3937654215425206, + "grad_norm": 0.0015239693457260728, + "learning_rate": 0.05691402957962713, + "loss": 0.07, + "num_input_tokens_seen": 36977536, + "step": 28525 + }, + { + "epoch": 1.394009723206215, + "grad_norm": 0.0006401727441698313, + "learning_rate": 0.05686784659471642, + "loss": 0.0914, + "num_input_tokens_seen": 36983872, + "step": 28530 + }, + { + "epoch": 1.3942540248699093, + "grad_norm": 0.0013899606419727206, + "learning_rate": 0.056821677971954136, + "loss": 0.0751, + "num_input_tokens_seen": 36990592, + "step": 28535 + }, + { + "epoch": 1.3944983265336037, + "grad_norm": 0.0010111293522641063, + "learning_rate": 0.05677552371846012, + "loss": 0.0782, + "num_input_tokens_seen": 36996864, + "step": 28540 + }, + { + "epoch": 1.394742628197298, + "grad_norm": 0.001128894160501659, + "learning_rate": 0.05672938384135182, + "loss": 0.0837, + "num_input_tokens_seen": 37003488, + "step": 28545 + }, + { + "epoch": 1.3949869298609925, + "grad_norm": 0.0009671600419096649, + "learning_rate": 0.05668325834774465, + "loss": 0.0981, + "num_input_tokens_seen": 37009760, + "step": 28550 + }, + { + "epoch": 1.3952312315246866, + "grad_norm": 0.0010645822621881962, + "learning_rate": 0.05663714724475177, + "loss": 0.1051, + "num_input_tokens_seen": 37016128, + "step": 28555 + }, + { + "epoch": 1.395475533188381, + "grad_norm": 0.0008958951220847666, + "learning_rate": 0.05659105053948403, + "loss": 0.0783, + "num_input_tokens_seen": 37022944, + "step": 28560 + }, + { + "epoch": 1.3957198348520754, + "grad_norm": 0.0018081923481076956, + "learning_rate": 0.056544968239050176, + "loss": 0.0868, + "num_input_tokens_seen": 37030368, + "step": 28565 + }, + { + "epoch": 1.3959641365157696, + "grad_norm": 0.0007862301426939666, + "learning_rate": 0.056498900350556616, + "loss": 0.073, + "num_input_tokens_seen": 37036704, + "step": 28570 + }, + { + "epoch": 1.396208438179464, + "grad_norm": 0.0013015479780733585, + "learning_rate": 0.05645284688110766, + "loss": 0.091, + "num_input_tokens_seen": 37042912, + "step": 28575 + }, + { + "epoch": 1.3964527398431583, + "grad_norm": 0.0010713490191847086, + "learning_rate": 0.05640680783780532, + "loss": 0.0882, + "num_input_tokens_seen": 37050048, + "step": 28580 + }, + { + "epoch": 1.3966970415068527, + "grad_norm": 0.0013394110137596726, + "learning_rate": 0.056360783227749324, + "loss": 0.0933, + "num_input_tokens_seen": 37056480, + "step": 28585 + }, + { + "epoch": 1.396941343170547, + "grad_norm": 0.0014353666920214891, + "learning_rate": 0.05631477305803728, + "loss": 0.0665, + "num_input_tokens_seen": 37062912, + "step": 28590 + }, + { + "epoch": 1.3971856448342415, + "grad_norm": 0.0013140442315489054, + "learning_rate": 0.05626877733576462, + "loss": 0.0837, + "num_input_tokens_seen": 37069504, + "step": 28595 + }, + { + "epoch": 1.3974299464979356, + "grad_norm": 0.000976457551587373, + "learning_rate": 0.05622279606802435, + "loss": 0.0693, + "num_input_tokens_seen": 37076064, + "step": 28600 + }, + { + "epoch": 1.3974299464979356, + "eval_loss": 0.0913688987493515, + "eval_runtime": 402.5635, + "eval_samples_per_second": 90.383, + "eval_steps_per_second": 22.598, + "num_input_tokens_seen": 37076064, + "step": 28600 + }, + { + "epoch": 1.39767424816163, + "grad_norm": 0.0015816559316590428, + "learning_rate": 0.05617682926190744, + "loss": 0.0902, + "num_input_tokens_seen": 37082496, + "step": 28605 + }, + { + "epoch": 1.3979185498253244, + "grad_norm": 0.0008354285964742303, + "learning_rate": 0.05613087692450248, + "loss": 0.0607, + "num_input_tokens_seen": 37089184, + "step": 28610 + }, + { + "epoch": 1.3981628514890185, + "grad_norm": 0.0004961327649652958, + "learning_rate": 0.05608493906289592, + "loss": 0.0779, + "num_input_tokens_seen": 37095424, + "step": 28615 + }, + { + "epoch": 1.398407153152713, + "grad_norm": 0.0007801969768479466, + "learning_rate": 0.05603901568417201, + "loss": 0.0824, + "num_input_tokens_seen": 37101792, + "step": 28620 + }, + { + "epoch": 1.3986514548164073, + "grad_norm": 0.0017786581302061677, + "learning_rate": 0.055993106795412625, + "loss": 0.092, + "num_input_tokens_seen": 37108096, + "step": 28625 + }, + { + "epoch": 1.3988957564801017, + "grad_norm": 0.0010716953547671437, + "learning_rate": 0.05594721240369759, + "loss": 0.0905, + "num_input_tokens_seen": 37114272, + "step": 28630 + }, + { + "epoch": 1.399140058143796, + "grad_norm": 0.0010557550704106688, + "learning_rate": 0.055901332516104296, + "loss": 0.0847, + "num_input_tokens_seen": 37120512, + "step": 28635 + }, + { + "epoch": 1.3993843598074902, + "grad_norm": 0.0010451392736285925, + "learning_rate": 0.05585546713970804, + "loss": 0.0659, + "num_input_tokens_seen": 37126976, + "step": 28640 + }, + { + "epoch": 1.3996286614711846, + "grad_norm": 0.0006678850622847676, + "learning_rate": 0.05580961628158189, + "loss": 0.0915, + "num_input_tokens_seen": 37133408, + "step": 28645 + }, + { + "epoch": 1.399872963134879, + "grad_norm": 0.0015024661552160978, + "learning_rate": 0.05576377994879659, + "loss": 0.099, + "num_input_tokens_seen": 37139712, + "step": 28650 + }, + { + "epoch": 1.4001172647985733, + "grad_norm": 0.0017856163904070854, + "learning_rate": 0.05571795814842063, + "loss": 0.0852, + "num_input_tokens_seen": 37146592, + "step": 28655 + }, + { + "epoch": 1.4003615664622675, + "grad_norm": 0.001268611871637404, + "learning_rate": 0.05567215088752037, + "loss": 0.0859, + "num_input_tokens_seen": 37153568, + "step": 28660 + }, + { + "epoch": 1.4006058681259619, + "grad_norm": 0.000752299907617271, + "learning_rate": 0.05562635817315981, + "loss": 0.0791, + "num_input_tokens_seen": 37160032, + "step": 28665 + }, + { + "epoch": 1.4008501697896563, + "grad_norm": 0.0011438884539529681, + "learning_rate": 0.05558058001240083, + "loss": 0.0838, + "num_input_tokens_seen": 37166560, + "step": 28670 + }, + { + "epoch": 1.4010944714533506, + "grad_norm": 0.0019745659083127975, + "learning_rate": 0.055534816412302915, + "loss": 0.0917, + "num_input_tokens_seen": 37172992, + "step": 28675 + }, + { + "epoch": 1.401338773117045, + "grad_norm": 0.0013442004565149546, + "learning_rate": 0.055489067379923436, + "loss": 0.0846, + "num_input_tokens_seen": 37179264, + "step": 28680 + }, + { + "epoch": 1.4015830747807392, + "grad_norm": 0.0016694579971954226, + "learning_rate": 0.055443332922317505, + "loss": 0.0705, + "num_input_tokens_seen": 37185856, + "step": 28685 + }, + { + "epoch": 1.4018273764444336, + "grad_norm": 0.001042093732394278, + "learning_rate": 0.055397613046537876, + "loss": 0.0682, + "num_input_tokens_seen": 37192320, + "step": 28690 + }, + { + "epoch": 1.402071678108128, + "grad_norm": 0.0015084448968991637, + "learning_rate": 0.055351907759635145, + "loss": 0.0933, + "num_input_tokens_seen": 37198944, + "step": 28695 + }, + { + "epoch": 1.4023159797718223, + "grad_norm": 0.0011908791493624449, + "learning_rate": 0.05530621706865772, + "loss": 0.0806, + "num_input_tokens_seen": 37205344, + "step": 28700 + }, + { + "epoch": 1.4025602814355165, + "grad_norm": 0.0012428310001268983, + "learning_rate": 0.055260540980651564, + "loss": 0.0676, + "num_input_tokens_seen": 37211488, + "step": 28705 + }, + { + "epoch": 1.4028045830992109, + "grad_norm": 0.001072256127372384, + "learning_rate": 0.05521487950266062, + "loss": 0.0879, + "num_input_tokens_seen": 37217536, + "step": 28710 + }, + { + "epoch": 1.4030488847629052, + "grad_norm": 0.0026098231319338083, + "learning_rate": 0.055169232641726344, + "loss": 0.1176, + "num_input_tokens_seen": 37223840, + "step": 28715 + }, + { + "epoch": 1.4032931864265996, + "grad_norm": 0.0016112910816445947, + "learning_rate": 0.055123600404888166, + "loss": 0.1117, + "num_input_tokens_seen": 37229888, + "step": 28720 + }, + { + "epoch": 1.403537488090294, + "grad_norm": 0.0013774132821708918, + "learning_rate": 0.05507798279918309, + "loss": 0.0641, + "num_input_tokens_seen": 37236896, + "step": 28725 + }, + { + "epoch": 1.4037817897539882, + "grad_norm": 0.0011911921901628375, + "learning_rate": 0.0550323798316459, + "loss": 0.0777, + "num_input_tokens_seen": 37243488, + "step": 28730 + }, + { + "epoch": 1.4040260914176825, + "grad_norm": 0.0014912665355950594, + "learning_rate": 0.05498679150930916, + "loss": 0.0755, + "num_input_tokens_seen": 37249952, + "step": 28735 + }, + { + "epoch": 1.404270393081377, + "grad_norm": 0.0010972392046824098, + "learning_rate": 0.05494121783920323, + "loss": 0.0817, + "num_input_tokens_seen": 37256736, + "step": 28740 + }, + { + "epoch": 1.4045146947450713, + "grad_norm": 0.0011140031274408102, + "learning_rate": 0.05489565882835605, + "loss": 0.0866, + "num_input_tokens_seen": 37263392, + "step": 28745 + }, + { + "epoch": 1.4047589964087654, + "grad_norm": 0.0007639821851626039, + "learning_rate": 0.05485011448379348, + "loss": 0.0733, + "num_input_tokens_seen": 37269504, + "step": 28750 + }, + { + "epoch": 1.4050032980724598, + "grad_norm": 0.0007793261320330203, + "learning_rate": 0.05480458481253893, + "loss": 0.0697, + "num_input_tokens_seen": 37276288, + "step": 28755 + }, + { + "epoch": 1.4052475997361542, + "grad_norm": 0.0008364845998585224, + "learning_rate": 0.054759069821613715, + "loss": 0.0824, + "num_input_tokens_seen": 37282816, + "step": 28760 + }, + { + "epoch": 1.4054919013998486, + "grad_norm": 0.0010390597162768245, + "learning_rate": 0.05471356951803683, + "loss": 0.0776, + "num_input_tokens_seen": 37289088, + "step": 28765 + }, + { + "epoch": 1.405736203063543, + "grad_norm": 0.0009602639474906027, + "learning_rate": 0.054668083908824945, + "loss": 0.0801, + "num_input_tokens_seen": 37295200, + "step": 28770 + }, + { + "epoch": 1.4059805047272371, + "grad_norm": 0.0014077703235670924, + "learning_rate": 0.054622613000992526, + "loss": 0.1059, + "num_input_tokens_seen": 37302016, + "step": 28775 + }, + { + "epoch": 1.4062248063909315, + "grad_norm": 0.0010705539025366306, + "learning_rate": 0.05457715680155182, + "loss": 0.0675, + "num_input_tokens_seen": 37308224, + "step": 28780 + }, + { + "epoch": 1.4064691080546259, + "grad_norm": 0.0011465270072221756, + "learning_rate": 0.05453171531751265, + "loss": 0.0821, + "num_input_tokens_seen": 37314528, + "step": 28785 + }, + { + "epoch": 1.4067134097183203, + "grad_norm": 0.0012098179431632161, + "learning_rate": 0.05448628855588276, + "loss": 0.0904, + "num_input_tokens_seen": 37320704, + "step": 28790 + }, + { + "epoch": 1.4069577113820144, + "grad_norm": 0.001341120689176023, + "learning_rate": 0.05444087652366746, + "loss": 0.0877, + "num_input_tokens_seen": 37326912, + "step": 28795 + }, + { + "epoch": 1.4072020130457088, + "grad_norm": 0.0012324274284765124, + "learning_rate": 0.05439547922786984, + "loss": 0.089, + "num_input_tokens_seen": 37333536, + "step": 28800 + }, + { + "epoch": 1.4072020130457088, + "eval_loss": 0.089190274477005, + "eval_runtime": 402.0917, + "eval_samples_per_second": 90.489, + "eval_steps_per_second": 22.624, + "num_input_tokens_seen": 37333536, + "step": 28800 + }, + { + "epoch": 1.4074463147094032, + "grad_norm": 0.0010906817624345422, + "learning_rate": 0.0543500966754908, + "loss": 0.0975, + "num_input_tokens_seen": 37340160, + "step": 28805 + }, + { + "epoch": 1.4076906163730976, + "grad_norm": 0.0008607955533079803, + "learning_rate": 0.05430472887352882, + "loss": 0.0748, + "num_input_tokens_seen": 37346720, + "step": 28810 + }, + { + "epoch": 1.407934918036792, + "grad_norm": 0.0015275586629286408, + "learning_rate": 0.05425937582898023, + "loss": 0.0956, + "num_input_tokens_seen": 37353152, + "step": 28815 + }, + { + "epoch": 1.408179219700486, + "grad_norm": 0.0015214981976896524, + "learning_rate": 0.054214037548839085, + "loss": 0.1048, + "num_input_tokens_seen": 37359232, + "step": 28820 + }, + { + "epoch": 1.4084235213641805, + "grad_norm": 0.0015678061172366142, + "learning_rate": 0.05416871404009703, + "loss": 0.0813, + "num_input_tokens_seen": 37365984, + "step": 28825 + }, + { + "epoch": 1.4086678230278749, + "grad_norm": 0.000832896854262799, + "learning_rate": 0.054123405309743605, + "loss": 0.0988, + "num_input_tokens_seen": 37372384, + "step": 28830 + }, + { + "epoch": 1.408912124691569, + "grad_norm": 0.00131369533482939, + "learning_rate": 0.0540781113647659, + "loss": 0.0732, + "num_input_tokens_seen": 37378784, + "step": 28835 + }, + { + "epoch": 1.4091564263552634, + "grad_norm": 0.0014107676688581705, + "learning_rate": 0.054032832212148836, + "loss": 0.0811, + "num_input_tokens_seen": 37385600, + "step": 28840 + }, + { + "epoch": 1.4094007280189578, + "grad_norm": 0.001004972611553967, + "learning_rate": 0.0539875678588751, + "loss": 0.0668, + "num_input_tokens_seen": 37392032, + "step": 28845 + }, + { + "epoch": 1.4096450296826522, + "grad_norm": 0.003159239422529936, + "learning_rate": 0.05394231831192492, + "loss": 0.0831, + "num_input_tokens_seen": 37398240, + "step": 28850 + }, + { + "epoch": 1.4098893313463465, + "grad_norm": 0.0010680241975933313, + "learning_rate": 0.05389708357827639, + "loss": 0.1061, + "num_input_tokens_seen": 37404864, + "step": 28855 + }, + { + "epoch": 1.410133633010041, + "grad_norm": 0.0013616137439385056, + "learning_rate": 0.05385186366490533, + "loss": 0.0724, + "num_input_tokens_seen": 37411776, + "step": 28860 + }, + { + "epoch": 1.410377934673735, + "grad_norm": 0.0017433097818866372, + "learning_rate": 0.053806658578785166, + "loss": 0.0764, + "num_input_tokens_seen": 37418048, + "step": 28865 + }, + { + "epoch": 1.4106222363374294, + "grad_norm": 0.001431377255357802, + "learning_rate": 0.05376146832688705, + "loss": 0.063, + "num_input_tokens_seen": 37424384, + "step": 28870 + }, + { + "epoch": 1.4108665380011238, + "grad_norm": 0.0008817470516078174, + "learning_rate": 0.053716292916179964, + "loss": 0.074, + "num_input_tokens_seen": 37431104, + "step": 28875 + }, + { + "epoch": 1.411110839664818, + "grad_norm": 0.0013118985807523131, + "learning_rate": 0.05367113235363045, + "loss": 0.0857, + "num_input_tokens_seen": 37437088, + "step": 28880 + }, + { + "epoch": 1.4113551413285124, + "grad_norm": 0.001891431980766356, + "learning_rate": 0.05362598664620289, + "loss": 0.0775, + "num_input_tokens_seen": 37443488, + "step": 28885 + }, + { + "epoch": 1.4115994429922067, + "grad_norm": 0.0017933498602360487, + "learning_rate": 0.053580855800859285, + "loss": 0.0978, + "num_input_tokens_seen": 37450176, + "step": 28890 + }, + { + "epoch": 1.4118437446559011, + "grad_norm": 0.0011845561675727367, + "learning_rate": 0.05353573982455938, + "loss": 0.0805, + "num_input_tokens_seen": 37456864, + "step": 28895 + }, + { + "epoch": 1.4120880463195955, + "grad_norm": 0.0008771494030952454, + "learning_rate": 0.053490638724260686, + "loss": 0.0598, + "num_input_tokens_seen": 37463232, + "step": 28900 + }, + { + "epoch": 1.4123323479832899, + "grad_norm": 0.0017830965807661414, + "learning_rate": 0.05344555250691827, + "loss": 0.096, + "num_input_tokens_seen": 37470016, + "step": 28905 + }, + { + "epoch": 1.412576649646984, + "grad_norm": 0.0009414932574145496, + "learning_rate": 0.053400481179485086, + "loss": 0.0843, + "num_input_tokens_seen": 37476352, + "step": 28910 + }, + { + "epoch": 1.4128209513106784, + "grad_norm": 0.001001955708488822, + "learning_rate": 0.05335542474891159, + "loss": 0.0619, + "num_input_tokens_seen": 37483136, + "step": 28915 + }, + { + "epoch": 1.4130652529743728, + "grad_norm": 0.0006363310967572033, + "learning_rate": 0.053310383222146124, + "loss": 0.0564, + "num_input_tokens_seen": 37489440, + "step": 28920 + }, + { + "epoch": 1.413309554638067, + "grad_norm": 0.0008382493397220969, + "learning_rate": 0.053265356606134684, + "loss": 0.0706, + "num_input_tokens_seen": 37496160, + "step": 28925 + }, + { + "epoch": 1.4135538563017613, + "grad_norm": 0.0017668287036940455, + "learning_rate": 0.053220344907820856, + "loss": 0.0842, + "num_input_tokens_seen": 37502304, + "step": 28930 + }, + { + "epoch": 1.4137981579654557, + "grad_norm": 0.00193500192835927, + "learning_rate": 0.05317534813414608, + "loss": 0.0859, + "num_input_tokens_seen": 37509248, + "step": 28935 + }, + { + "epoch": 1.41404245962915, + "grad_norm": 0.001387807191349566, + "learning_rate": 0.05313036629204942, + "loss": 0.0863, + "num_input_tokens_seen": 37515392, + "step": 28940 + }, + { + "epoch": 1.4142867612928445, + "grad_norm": 0.0012462519807741046, + "learning_rate": 0.05308539938846756, + "loss": 0.0835, + "num_input_tokens_seen": 37521824, + "step": 28945 + }, + { + "epoch": 1.4145310629565389, + "grad_norm": 0.001364604220725596, + "learning_rate": 0.05304044743033507, + "loss": 0.0697, + "num_input_tokens_seen": 37528896, + "step": 28950 + }, + { + "epoch": 1.414775364620233, + "grad_norm": 0.001081672846339643, + "learning_rate": 0.05299551042458401, + "loss": 0.0891, + "num_input_tokens_seen": 37534944, + "step": 28955 + }, + { + "epoch": 1.4150196662839274, + "grad_norm": 0.0012210081331431866, + "learning_rate": 0.052950588378144266, + "loss": 0.0995, + "num_input_tokens_seen": 37541280, + "step": 28960 + }, + { + "epoch": 1.4152639679476218, + "grad_norm": 0.0011568287154659629, + "learning_rate": 0.052905681297943465, + "loss": 0.0938, + "num_input_tokens_seen": 37548000, + "step": 28965 + }, + { + "epoch": 1.415508269611316, + "grad_norm": 0.0010846139630302787, + "learning_rate": 0.0528607891909067, + "loss": 0.0922, + "num_input_tokens_seen": 37553984, + "step": 28970 + }, + { + "epoch": 1.4157525712750103, + "grad_norm": 0.0012585598742589355, + "learning_rate": 0.05281591206395697, + "loss": 0.0802, + "num_input_tokens_seen": 37560288, + "step": 28975 + }, + { + "epoch": 1.4159968729387047, + "grad_norm": 0.002088769106194377, + "learning_rate": 0.05277104992401496, + "loss": 0.0985, + "num_input_tokens_seen": 37567360, + "step": 28980 + }, + { + "epoch": 1.416241174602399, + "grad_norm": 0.0012534359702840447, + "learning_rate": 0.05272620277799884, + "loss": 0.1254, + "num_input_tokens_seen": 37573696, + "step": 28985 + }, + { + "epoch": 1.4164854762660934, + "grad_norm": 0.003071385435760021, + "learning_rate": 0.05268137063282473, + "loss": 0.1018, + "num_input_tokens_seen": 37580256, + "step": 28990 + }, + { + "epoch": 1.4167297779297878, + "grad_norm": 0.0008603695314377546, + "learning_rate": 0.0526365534954062, + "loss": 0.0838, + "num_input_tokens_seen": 37586784, + "step": 28995 + }, + { + "epoch": 1.416974079593482, + "grad_norm": 0.001182161970064044, + "learning_rate": 0.052591751372654656, + "loss": 0.0607, + "num_input_tokens_seen": 37593216, + "step": 29000 + }, + { + "epoch": 1.416974079593482, + "eval_loss": 0.09011533856391907, + "eval_runtime": 402.7839, + "eval_samples_per_second": 90.334, + "eval_steps_per_second": 22.585, + "num_input_tokens_seen": 37593216, + "step": 29000 + }, + { + "epoch": 1.4172183812571764, + "grad_norm": 0.0015873395605012774, + "learning_rate": 0.05254696427147921, + "loss": 0.0849, + "num_input_tokens_seen": 37599232, + "step": 29005 + }, + { + "epoch": 1.4174626829208707, + "grad_norm": 0.0014863209798932076, + "learning_rate": 0.052502192198786546, + "loss": 0.0978, + "num_input_tokens_seen": 37605504, + "step": 29010 + }, + { + "epoch": 1.417706984584565, + "grad_norm": 0.0014119843253865838, + "learning_rate": 0.05245743516148103, + "loss": 0.0738, + "num_input_tokens_seen": 37612192, + "step": 29015 + }, + { + "epoch": 1.4179512862482593, + "grad_norm": 0.0007289907080121338, + "learning_rate": 0.05241269316646486, + "loss": 0.0855, + "num_input_tokens_seen": 37618688, + "step": 29020 + }, + { + "epoch": 1.4181955879119537, + "grad_norm": 0.0014910363825038075, + "learning_rate": 0.052367966220637725, + "loss": 0.0862, + "num_input_tokens_seen": 37625312, + "step": 29025 + }, + { + "epoch": 1.418439889575648, + "grad_norm": 0.0010491793509572744, + "learning_rate": 0.05232325433089716, + "loss": 0.0789, + "num_input_tokens_seen": 37631328, + "step": 29030 + }, + { + "epoch": 1.4186841912393424, + "grad_norm": 0.0008000056259334087, + "learning_rate": 0.052278557504138214, + "loss": 0.0772, + "num_input_tokens_seen": 37637536, + "step": 29035 + }, + { + "epoch": 1.4189284929030368, + "grad_norm": 0.0017371276626363397, + "learning_rate": 0.05223387574725372, + "loss": 0.0835, + "num_input_tokens_seen": 37643584, + "step": 29040 + }, + { + "epoch": 1.419172794566731, + "grad_norm": 0.0012135491706430912, + "learning_rate": 0.05218920906713428, + "loss": 0.0735, + "num_input_tokens_seen": 37649920, + "step": 29045 + }, + { + "epoch": 1.4194170962304253, + "grad_norm": 0.001636622822843492, + "learning_rate": 0.05214455747066789, + "loss": 0.1105, + "num_input_tokens_seen": 37656672, + "step": 29050 + }, + { + "epoch": 1.4196613978941197, + "grad_norm": 0.001160430838353932, + "learning_rate": 0.05209992096474048, + "loss": 0.0709, + "num_input_tokens_seen": 37663328, + "step": 29055 + }, + { + "epoch": 1.4199056995578139, + "grad_norm": 0.0009640877251513302, + "learning_rate": 0.05205529955623559, + "loss": 0.0733, + "num_input_tokens_seen": 37669824, + "step": 29060 + }, + { + "epoch": 1.4201500012215083, + "grad_norm": 0.0017413304885849357, + "learning_rate": 0.052010693252034314, + "loss": 0.088, + "num_input_tokens_seen": 37675872, + "step": 29065 + }, + { + "epoch": 1.4203943028852026, + "grad_norm": 0.0006355468649417162, + "learning_rate": 0.0519661020590156, + "loss": 0.0976, + "num_input_tokens_seen": 37682464, + "step": 29070 + }, + { + "epoch": 1.420638604548897, + "grad_norm": 0.0010838659945875406, + "learning_rate": 0.05192152598405586, + "loss": 0.0877, + "num_input_tokens_seen": 37688800, + "step": 29075 + }, + { + "epoch": 1.4208829062125914, + "grad_norm": 0.0013871613191440701, + "learning_rate": 0.05187696503402941, + "loss": 0.0927, + "num_input_tokens_seen": 37695936, + "step": 29080 + }, + { + "epoch": 1.4211272078762858, + "grad_norm": 0.0008756861207075417, + "learning_rate": 0.05183241921580798, + "loss": 0.0612, + "num_input_tokens_seen": 37702560, + "step": 29085 + }, + { + "epoch": 1.42137150953998, + "grad_norm": 0.0015643289079889655, + "learning_rate": 0.051787888536261206, + "loss": 0.09, + "num_input_tokens_seen": 37708960, + "step": 29090 + }, + { + "epoch": 1.4216158112036743, + "grad_norm": 0.0008652537944726646, + "learning_rate": 0.051743373002256184, + "loss": 0.0904, + "num_input_tokens_seen": 37715712, + "step": 29095 + }, + { + "epoch": 1.4218601128673687, + "grad_norm": 0.0009192412253469229, + "learning_rate": 0.05169887262065787, + "loss": 0.0758, + "num_input_tokens_seen": 37722208, + "step": 29100 + }, + { + "epoch": 1.4221044145310628, + "grad_norm": 0.0013153868494555354, + "learning_rate": 0.051654387398328665, + "loss": 0.0866, + "num_input_tokens_seen": 37728640, + "step": 29105 + }, + { + "epoch": 1.4223487161947572, + "grad_norm": 0.0010991364251822233, + "learning_rate": 0.05160991734212888, + "loss": 0.0981, + "num_input_tokens_seen": 37735008, + "step": 29110 + }, + { + "epoch": 1.4225930178584516, + "grad_norm": 0.0009213042794726789, + "learning_rate": 0.051565462458916224, + "loss": 0.0668, + "num_input_tokens_seen": 37741216, + "step": 29115 + }, + { + "epoch": 1.422837319522146, + "grad_norm": 0.0009734188788570464, + "learning_rate": 0.05152102275554627, + "loss": 0.0774, + "num_input_tokens_seen": 37747648, + "step": 29120 + }, + { + "epoch": 1.4230816211858404, + "grad_norm": 0.001353623578324914, + "learning_rate": 0.05147659823887222, + "loss": 0.0891, + "num_input_tokens_seen": 37754240, + "step": 29125 + }, + { + "epoch": 1.4233259228495347, + "grad_norm": 0.0012764108832925558, + "learning_rate": 0.05143218891574479, + "loss": 0.0598, + "num_input_tokens_seen": 37760640, + "step": 29130 + }, + { + "epoch": 1.423570224513229, + "grad_norm": 0.0013018515892326832, + "learning_rate": 0.0513877947930125, + "loss": 0.0635, + "num_input_tokens_seen": 37767200, + "step": 29135 + }, + { + "epoch": 1.4238145261769233, + "grad_norm": 0.0016073302831500769, + "learning_rate": 0.051343415877521566, + "loss": 0.1145, + "num_input_tokens_seen": 37773376, + "step": 29140 + }, + { + "epoch": 1.4240588278406177, + "grad_norm": 0.0014255531132221222, + "learning_rate": 0.051299052176115634, + "loss": 0.0731, + "num_input_tokens_seen": 37780320, + "step": 29145 + }, + { + "epoch": 1.4243031295043118, + "grad_norm": 0.0010886286618188024, + "learning_rate": 0.051254703695636256, + "loss": 0.0829, + "num_input_tokens_seen": 37787200, + "step": 29150 + }, + { + "epoch": 1.4245474311680062, + "grad_norm": 0.0008470896864309907, + "learning_rate": 0.05121037044292249, + "loss": 0.093, + "num_input_tokens_seen": 37793856, + "step": 29155 + }, + { + "epoch": 1.4247917328317006, + "grad_norm": 0.0011618374846875668, + "learning_rate": 0.05116605242481101, + "loss": 0.0711, + "num_input_tokens_seen": 37799840, + "step": 29160 + }, + { + "epoch": 1.425036034495395, + "grad_norm": 0.0011980005074292421, + "learning_rate": 0.05112174964813634, + "loss": 0.0982, + "num_input_tokens_seen": 37806304, + "step": 29165 + }, + { + "epoch": 1.4252803361590893, + "grad_norm": 0.0012856371467933059, + "learning_rate": 0.05107746211973038, + "loss": 0.0687, + "num_input_tokens_seen": 37812576, + "step": 29170 + }, + { + "epoch": 1.4255246378227835, + "grad_norm": 0.0011582098668441176, + "learning_rate": 0.05103318984642291, + "loss": 0.0625, + "num_input_tokens_seen": 37819296, + "step": 29175 + }, + { + "epoch": 1.4257689394864779, + "grad_norm": 0.0014885020209476352, + "learning_rate": 0.05098893283504131, + "loss": 0.063, + "num_input_tokens_seen": 37825504, + "step": 29180 + }, + { + "epoch": 1.4260132411501723, + "grad_norm": 0.0016818542499095201, + "learning_rate": 0.050944691092410475, + "loss": 0.0935, + "num_input_tokens_seen": 37831808, + "step": 29185 + }, + { + "epoch": 1.4262575428138666, + "grad_norm": 0.001049654558300972, + "learning_rate": 0.05090046462535313, + "loss": 0.0879, + "num_input_tokens_seen": 37838144, + "step": 29190 + }, + { + "epoch": 1.4265018444775608, + "grad_norm": 0.0013904717052355409, + "learning_rate": 0.050856253440689454, + "loss": 0.0904, + "num_input_tokens_seen": 37844384, + "step": 29195 + }, + { + "epoch": 1.4267461461412552, + "grad_norm": 0.0011136592365801334, + "learning_rate": 0.050812057545237405, + "loss": 0.0662, + "num_input_tokens_seen": 37850816, + "step": 29200 + }, + { + "epoch": 1.4267461461412552, + "eval_loss": 0.08789268136024475, + "eval_runtime": 402.2948, + "eval_samples_per_second": 90.444, + "eval_steps_per_second": 22.613, + "num_input_tokens_seen": 37850816, + "step": 29200 + }, + { + "epoch": 1.4269904478049495, + "grad_norm": 0.0008230743114836514, + "learning_rate": 0.0507678769458126, + "loss": 0.0801, + "num_input_tokens_seen": 37857184, + "step": 29205 + }, + { + "epoch": 1.427234749468644, + "grad_norm": 0.0009686574339866638, + "learning_rate": 0.050723711649228155, + "loss": 0.0734, + "num_input_tokens_seen": 37863520, + "step": 29210 + }, + { + "epoch": 1.4274790511323383, + "grad_norm": 0.001553932554088533, + "learning_rate": 0.05067956166229496, + "loss": 0.1091, + "num_input_tokens_seen": 37869536, + "step": 29215 + }, + { + "epoch": 1.4277233527960325, + "grad_norm": 0.0014272663975134492, + "learning_rate": 0.05063542699182155, + "loss": 0.0584, + "num_input_tokens_seen": 37875936, + "step": 29220 + }, + { + "epoch": 1.4279676544597268, + "grad_norm": 0.0019744429737329483, + "learning_rate": 0.050591307644613996, + "loss": 0.1029, + "num_input_tokens_seen": 37882880, + "step": 29225 + }, + { + "epoch": 1.4282119561234212, + "grad_norm": 0.002485546749085188, + "learning_rate": 0.05054720362747599, + "loss": 0.1026, + "num_input_tokens_seen": 37889440, + "step": 29230 + }, + { + "epoch": 1.4284562577871156, + "grad_norm": 0.001700089080259204, + "learning_rate": 0.050503114947209035, + "loss": 0.0852, + "num_input_tokens_seen": 37895936, + "step": 29235 + }, + { + "epoch": 1.4287005594508098, + "grad_norm": 0.0013177086366340518, + "learning_rate": 0.05045904161061207, + "loss": 0.0693, + "num_input_tokens_seen": 37902752, + "step": 29240 + }, + { + "epoch": 1.4289448611145041, + "grad_norm": 0.0018559419550001621, + "learning_rate": 0.05041498362448185, + "loss": 0.0959, + "num_input_tokens_seen": 37909376, + "step": 29245 + }, + { + "epoch": 1.4291891627781985, + "grad_norm": 0.001065060612745583, + "learning_rate": 0.05037094099561256, + "loss": 0.0723, + "num_input_tokens_seen": 37915904, + "step": 29250 + }, + { + "epoch": 1.429433464441893, + "grad_norm": 0.002471179934218526, + "learning_rate": 0.05032691373079624, + "loss": 0.0882, + "num_input_tokens_seen": 37922784, + "step": 29255 + }, + { + "epoch": 1.4296777661055873, + "grad_norm": 0.001213450450450182, + "learning_rate": 0.05028290183682234, + "loss": 0.0771, + "num_input_tokens_seen": 37928960, + "step": 29260 + }, + { + "epoch": 1.4299220677692814, + "grad_norm": 0.0017120590200647712, + "learning_rate": 0.050238905320478096, + "loss": 0.0815, + "num_input_tokens_seen": 37935776, + "step": 29265 + }, + { + "epoch": 1.4301663694329758, + "grad_norm": 0.0016711890930309892, + "learning_rate": 0.05019492418854838, + "loss": 0.0845, + "num_input_tokens_seen": 37941920, + "step": 29270 + }, + { + "epoch": 1.4304106710966702, + "grad_norm": 0.0012134760618209839, + "learning_rate": 0.05015095844781554, + "loss": 0.0757, + "num_input_tokens_seen": 37948736, + "step": 29275 + }, + { + "epoch": 1.4306549727603646, + "grad_norm": 0.0017868392169475555, + "learning_rate": 0.05010700810505968, + "loss": 0.0929, + "num_input_tokens_seen": 37954816, + "step": 29280 + }, + { + "epoch": 1.4308992744240587, + "grad_norm": 0.0007005748921073973, + "learning_rate": 0.05006307316705856, + "loss": 0.0627, + "num_input_tokens_seen": 37961536, + "step": 29285 + }, + { + "epoch": 1.431143576087753, + "grad_norm": 0.0014874041080474854, + "learning_rate": 0.0500191536405874, + "loss": 0.0823, + "num_input_tokens_seen": 37967968, + "step": 29290 + }, + { + "epoch": 1.4313878777514475, + "grad_norm": 0.0008234120323322713, + "learning_rate": 0.04997524953241922, + "loss": 0.0568, + "num_input_tokens_seen": 37974624, + "step": 29295 + }, + { + "epoch": 1.4316321794151419, + "grad_norm": 0.0012747971341013908, + "learning_rate": 0.049931360849324556, + "loss": 0.0648, + "num_input_tokens_seen": 37980576, + "step": 29300 + }, + { + "epoch": 1.4318764810788362, + "grad_norm": 0.002228298457339406, + "learning_rate": 0.04988748759807155, + "loss": 0.0988, + "num_input_tokens_seen": 37987136, + "step": 29305 + }, + { + "epoch": 1.4321207827425304, + "grad_norm": 0.002180425450205803, + "learning_rate": 0.0498436297854261, + "loss": 0.1155, + "num_input_tokens_seen": 37993344, + "step": 29310 + }, + { + "epoch": 1.4323650844062248, + "grad_norm": 0.001184134162031114, + "learning_rate": 0.04979978741815152, + "loss": 0.0979, + "num_input_tokens_seen": 38000000, + "step": 29315 + }, + { + "epoch": 1.4326093860699192, + "grad_norm": 0.001642123213969171, + "learning_rate": 0.04975596050300891, + "loss": 0.0778, + "num_input_tokens_seen": 38006464, + "step": 29320 + }, + { + "epoch": 1.4328536877336135, + "grad_norm": 0.0014475476928055286, + "learning_rate": 0.049712149046757005, + "loss": 0.0709, + "num_input_tokens_seen": 38012768, + "step": 29325 + }, + { + "epoch": 1.4330979893973077, + "grad_norm": 0.0010778657160699368, + "learning_rate": 0.04966835305615194, + "loss": 0.0831, + "num_input_tokens_seen": 38019136, + "step": 29330 + }, + { + "epoch": 1.433342291061002, + "grad_norm": 0.0014726684894412756, + "learning_rate": 0.049624572537947755, + "loss": 0.0873, + "num_input_tokens_seen": 38025696, + "step": 29335 + }, + { + "epoch": 1.4335865927246965, + "grad_norm": 0.0017015283228829503, + "learning_rate": 0.04958080749889582, + "loss": 0.0907, + "num_input_tokens_seen": 38032192, + "step": 29340 + }, + { + "epoch": 1.4338308943883908, + "grad_norm": 0.0008359932107850909, + "learning_rate": 0.049537057945745304, + "loss": 0.0897, + "num_input_tokens_seen": 38038816, + "step": 29345 + }, + { + "epoch": 1.4340751960520852, + "grad_norm": 0.00128098763525486, + "learning_rate": 0.049493323885243, + "loss": 0.0863, + "num_input_tokens_seen": 38045216, + "step": 29350 + }, + { + "epoch": 1.4343194977157794, + "grad_norm": 0.0010825099889189005, + "learning_rate": 0.04944960532413318, + "loss": 0.0627, + "num_input_tokens_seen": 38051584, + "step": 29355 + }, + { + "epoch": 1.4345637993794738, + "grad_norm": 0.0015671361470595002, + "learning_rate": 0.049405902269157774, + "loss": 0.0881, + "num_input_tokens_seen": 38058080, + "step": 29360 + }, + { + "epoch": 1.4348081010431681, + "grad_norm": 0.0008425941341556609, + "learning_rate": 0.04936221472705646, + "loss": 0.0822, + "num_input_tokens_seen": 38066112, + "step": 29365 + }, + { + "epoch": 1.4350524027068623, + "grad_norm": 0.0012435941025614738, + "learning_rate": 0.04931854270456632, + "loss": 0.0648, + "num_input_tokens_seen": 38072416, + "step": 29370 + }, + { + "epoch": 1.4352967043705567, + "grad_norm": 0.0018774218624457717, + "learning_rate": 0.049274886208422075, + "loss": 0.1047, + "num_input_tokens_seen": 38079104, + "step": 29375 + }, + { + "epoch": 1.435541006034251, + "grad_norm": 0.000992129324004054, + "learning_rate": 0.049231245245356235, + "loss": 0.0671, + "num_input_tokens_seen": 38085568, + "step": 29380 + }, + { + "epoch": 1.4357853076979454, + "grad_norm": 0.0012053243117406964, + "learning_rate": 0.049187619822098655, + "loss": 0.1005, + "num_input_tokens_seen": 38091904, + "step": 29385 + }, + { + "epoch": 1.4360296093616398, + "grad_norm": 0.0017044511623680592, + "learning_rate": 0.04914400994537705, + "loss": 0.0676, + "num_input_tokens_seen": 38098624, + "step": 29390 + }, + { + "epoch": 1.4362739110253342, + "grad_norm": 0.0017142101423814893, + "learning_rate": 0.049100415621916485, + "loss": 0.0846, + "num_input_tokens_seen": 38104896, + "step": 29395 + }, + { + "epoch": 1.4365182126890284, + "grad_norm": 0.001973492093384266, + "learning_rate": 0.04905683685843981, + "loss": 0.0756, + "num_input_tokens_seen": 38111232, + "step": 29400 + }, + { + "epoch": 1.4365182126890284, + "eval_loss": 0.08719221502542496, + "eval_runtime": 402.3968, + "eval_samples_per_second": 90.421, + "eval_steps_per_second": 22.607, + "num_input_tokens_seen": 38111232, + "step": 29400 + }, + { + "epoch": 1.4367625143527227, + "grad_norm": 0.000654537754599005, + "learning_rate": 0.049013273661667495, + "loss": 0.0659, + "num_input_tokens_seen": 38117632, + "step": 29405 + }, + { + "epoch": 1.437006816016417, + "grad_norm": 0.001384738483466208, + "learning_rate": 0.048969726038317396, + "loss": 0.0864, + "num_input_tokens_seen": 38124544, + "step": 29410 + }, + { + "epoch": 1.4372511176801113, + "grad_norm": 0.0010680003324523568, + "learning_rate": 0.048926193995105206, + "loss": 0.0743, + "num_input_tokens_seen": 38130400, + "step": 29415 + }, + { + "epoch": 1.4374954193438056, + "grad_norm": 0.0012797893723472953, + "learning_rate": 0.048882677538744035, + "loss": 0.1073, + "num_input_tokens_seen": 38136704, + "step": 29420 + }, + { + "epoch": 1.4377397210075, + "grad_norm": 0.00095588737167418, + "learning_rate": 0.048839176675944715, + "loss": 0.0598, + "num_input_tokens_seen": 38143488, + "step": 29425 + }, + { + "epoch": 1.4379840226711944, + "grad_norm": 0.0015259679639711976, + "learning_rate": 0.04879569141341566, + "loss": 0.0965, + "num_input_tokens_seen": 38149856, + "step": 29430 + }, + { + "epoch": 1.4382283243348888, + "grad_norm": 0.0011113915825262666, + "learning_rate": 0.04875222175786274, + "loss": 0.1153, + "num_input_tokens_seen": 38156000, + "step": 29435 + }, + { + "epoch": 1.4384726259985832, + "grad_norm": 0.0018464884487912059, + "learning_rate": 0.04870876771598966, + "loss": 0.0723, + "num_input_tokens_seen": 38162656, + "step": 29440 + }, + { + "epoch": 1.4387169276622773, + "grad_norm": 0.0008861309033818543, + "learning_rate": 0.04866532929449744, + "loss": 0.0927, + "num_input_tokens_seen": 38170016, + "step": 29445 + }, + { + "epoch": 1.4389612293259717, + "grad_norm": 0.0015494243707507849, + "learning_rate": 0.048621906500084945, + "loss": 0.0769, + "num_input_tokens_seen": 38176448, + "step": 29450 + }, + { + "epoch": 1.439205530989666, + "grad_norm": 0.0015705206897109747, + "learning_rate": 0.04857849933944845, + "loss": 0.096, + "num_input_tokens_seen": 38182560, + "step": 29455 + }, + { + "epoch": 1.4394498326533602, + "grad_norm": 0.0021756382193416357, + "learning_rate": 0.048535107819281866, + "loss": 0.0939, + "num_input_tokens_seen": 38189344, + "step": 29460 + }, + { + "epoch": 1.4396941343170546, + "grad_norm": 0.0008074399665929377, + "learning_rate": 0.04849173194627675, + "loss": 0.0609, + "num_input_tokens_seen": 38196032, + "step": 29465 + }, + { + "epoch": 1.439938435980749, + "grad_norm": 0.0014314927393570542, + "learning_rate": 0.04844837172712223, + "loss": 0.0689, + "num_input_tokens_seen": 38202752, + "step": 29470 + }, + { + "epoch": 1.4401827376444434, + "grad_norm": 0.0016502897487953305, + "learning_rate": 0.04840502716850494, + "loss": 0.077, + "num_input_tokens_seen": 38209376, + "step": 29475 + }, + { + "epoch": 1.4404270393081378, + "grad_norm": 0.0015628064284101129, + "learning_rate": 0.04836169827710916, + "loss": 0.0858, + "num_input_tokens_seen": 38215584, + "step": 29480 + }, + { + "epoch": 1.4406713409718321, + "grad_norm": 0.0016271424246951938, + "learning_rate": 0.04831838505961684, + "loss": 0.0798, + "num_input_tokens_seen": 38222048, + "step": 29485 + }, + { + "epoch": 1.4409156426355263, + "grad_norm": 0.0009120592148974538, + "learning_rate": 0.048275087522707295, + "loss": 0.0999, + "num_input_tokens_seen": 38228800, + "step": 29490 + }, + { + "epoch": 1.4411599442992207, + "grad_norm": 0.0008434168412350118, + "learning_rate": 0.04823180567305766, + "loss": 0.0488, + "num_input_tokens_seen": 38235840, + "step": 29495 + }, + { + "epoch": 1.441404245962915, + "grad_norm": 0.00199145614169538, + "learning_rate": 0.04818853951734244, + "loss": 0.0861, + "num_input_tokens_seen": 38242400, + "step": 29500 + }, + { + "epoch": 1.4416485476266092, + "grad_norm": 0.0017326996894553304, + "learning_rate": 0.04814528906223387, + "loss": 0.0991, + "num_input_tokens_seen": 38248352, + "step": 29505 + }, + { + "epoch": 1.4418928492903036, + "grad_norm": 0.0012503588804975152, + "learning_rate": 0.04810205431440177, + "loss": 0.0538, + "num_input_tokens_seen": 38254496, + "step": 29510 + }, + { + "epoch": 1.442137150953998, + "grad_norm": 0.000965778308454901, + "learning_rate": 0.04805883528051341, + "loss": 0.0838, + "num_input_tokens_seen": 38260768, + "step": 29515 + }, + { + "epoch": 1.4423814526176924, + "grad_norm": 0.0007693669758737087, + "learning_rate": 0.048015631967233685, + "loss": 0.0827, + "num_input_tokens_seen": 38267456, + "step": 29520 + }, + { + "epoch": 1.4426257542813867, + "grad_norm": 0.0015777373919263482, + "learning_rate": 0.04797244438122517, + "loss": 0.1203, + "num_input_tokens_seen": 38273856, + "step": 29525 + }, + { + "epoch": 1.442870055945081, + "grad_norm": 0.0011354215675964952, + "learning_rate": 0.04792927252914784, + "loss": 0.0794, + "num_input_tokens_seen": 38280224, + "step": 29530 + }, + { + "epoch": 1.4431143576087753, + "grad_norm": 0.0013238011160865426, + "learning_rate": 0.04788611641765944, + "loss": 0.1016, + "num_input_tokens_seen": 38286432, + "step": 29535 + }, + { + "epoch": 1.4433586592724696, + "grad_norm": 0.0010105998953804374, + "learning_rate": 0.04784297605341508, + "loss": 0.0892, + "num_input_tokens_seen": 38293248, + "step": 29540 + }, + { + "epoch": 1.443602960936164, + "grad_norm": 0.001359340501949191, + "learning_rate": 0.04779985144306761, + "loss": 0.0631, + "num_input_tokens_seen": 38299520, + "step": 29545 + }, + { + "epoch": 1.4438472625998582, + "grad_norm": 0.0009825152810662985, + "learning_rate": 0.047756742593267405, + "loss": 0.089, + "num_input_tokens_seen": 38305728, + "step": 29550 + }, + { + "epoch": 1.4440915642635526, + "grad_norm": 0.0011040950194001198, + "learning_rate": 0.047713649510662315, + "loss": 0.0728, + "num_input_tokens_seen": 38312544, + "step": 29555 + }, + { + "epoch": 1.444335865927247, + "grad_norm": 0.0006762828561477363, + "learning_rate": 0.04767057220189789, + "loss": 0.0819, + "num_input_tokens_seen": 38318432, + "step": 29560 + }, + { + "epoch": 1.4445801675909413, + "grad_norm": 0.0009768622694537044, + "learning_rate": 0.04762751067361722, + "loss": 0.0707, + "num_input_tokens_seen": 38324896, + "step": 29565 + }, + { + "epoch": 1.4448244692546357, + "grad_norm": 0.0013375330017879605, + "learning_rate": 0.04758446493246086, + "loss": 0.0737, + "num_input_tokens_seen": 38331360, + "step": 29570 + }, + { + "epoch": 1.44506877091833, + "grad_norm": 0.0019676857627928257, + "learning_rate": 0.047541434985067084, + "loss": 0.0816, + "num_input_tokens_seen": 38337216, + "step": 29575 + }, + { + "epoch": 1.4453130725820242, + "grad_norm": 0.0011283357162028551, + "learning_rate": 0.047498420838071556, + "loss": 0.0499, + "num_input_tokens_seen": 38343680, + "step": 29580 + }, + { + "epoch": 1.4455573742457186, + "grad_norm": 0.001984336879104376, + "learning_rate": 0.04745542249810772, + "loss": 0.0939, + "num_input_tokens_seen": 38350176, + "step": 29585 + }, + { + "epoch": 1.445801675909413, + "grad_norm": 0.0014612541999667883, + "learning_rate": 0.047412439971806324, + "loss": 0.0844, + "num_input_tokens_seen": 38356512, + "step": 29590 + }, + { + "epoch": 1.4460459775731072, + "grad_norm": 0.0018713921308517456, + "learning_rate": 0.04736947326579592, + "loss": 0.0865, + "num_input_tokens_seen": 38362944, + "step": 29595 + }, + { + "epoch": 1.4462902792368015, + "grad_norm": 0.001011454383842647, + "learning_rate": 0.04732652238670245, + "loss": 0.0639, + "num_input_tokens_seen": 38370144, + "step": 29600 + }, + { + "epoch": 1.4462902792368015, + "eval_loss": 0.0867699384689331, + "eval_runtime": 402.3036, + "eval_samples_per_second": 90.442, + "eval_steps_per_second": 22.612, + "num_input_tokens_seen": 38370144, + "step": 29600 + }, + { + "epoch": 1.446534580900496, + "grad_norm": 0.0016825450584292412, + "learning_rate": 0.04728358734114952, + "loss": 0.0854, + "num_input_tokens_seen": 38376416, + "step": 29605 + }, + { + "epoch": 1.4467788825641903, + "grad_norm": 0.0032600299455225468, + "learning_rate": 0.04724066813575821, + "loss": 0.0957, + "num_input_tokens_seen": 38382976, + "step": 29610 + }, + { + "epoch": 1.4470231842278847, + "grad_norm": 0.0014635034603998065, + "learning_rate": 0.04719776477714729, + "loss": 0.0876, + "num_input_tokens_seen": 38389472, + "step": 29615 + }, + { + "epoch": 1.447267485891579, + "grad_norm": 0.0005662399344146252, + "learning_rate": 0.047154877271932856, + "loss": 0.0691, + "num_input_tokens_seen": 38396640, + "step": 29620 + }, + { + "epoch": 1.4475117875552732, + "grad_norm": 0.0020705279894173145, + "learning_rate": 0.0471120056267288, + "loss": 0.0947, + "num_input_tokens_seen": 38403360, + "step": 29625 + }, + { + "epoch": 1.4477560892189676, + "grad_norm": 0.0017746041994541883, + "learning_rate": 0.047069149848146495, + "loss": 0.1483, + "num_input_tokens_seen": 38409504, + "step": 29630 + }, + { + "epoch": 1.448000390882662, + "grad_norm": 0.0006633977172896266, + "learning_rate": 0.04702630994279473, + "loss": 0.065, + "num_input_tokens_seen": 38415872, + "step": 29635 + }, + { + "epoch": 1.4482446925463561, + "grad_norm": 0.0014810521388426423, + "learning_rate": 0.046983485917280035, + "loss": 0.0649, + "num_input_tokens_seen": 38422304, + "step": 29640 + }, + { + "epoch": 1.4484889942100505, + "grad_norm": 0.0010765662882477045, + "learning_rate": 0.04694067777820644, + "loss": 0.089, + "num_input_tokens_seen": 38428672, + "step": 29645 + }, + { + "epoch": 1.4487332958737449, + "grad_norm": 0.0012070094235241413, + "learning_rate": 0.046897885532175415, + "loss": 0.0766, + "num_input_tokens_seen": 38434784, + "step": 29650 + }, + { + "epoch": 1.4489775975374393, + "grad_norm": 0.0016397691797465086, + "learning_rate": 0.04685510918578613, + "loss": 0.0819, + "num_input_tokens_seen": 38441152, + "step": 29655 + }, + { + "epoch": 1.4492218992011336, + "grad_norm": 0.0014293675776571035, + "learning_rate": 0.04681234874563519, + "loss": 0.0769, + "num_input_tokens_seen": 38447872, + "step": 29660 + }, + { + "epoch": 1.449466200864828, + "grad_norm": 0.0015556605067104101, + "learning_rate": 0.046769604218316836, + "loss": 0.0832, + "num_input_tokens_seen": 38454208, + "step": 29665 + }, + { + "epoch": 1.4497105025285222, + "grad_norm": 0.0008524667355231941, + "learning_rate": 0.04672687561042279, + "loss": 0.0855, + "num_input_tokens_seen": 38460640, + "step": 29670 + }, + { + "epoch": 1.4499548041922166, + "grad_norm": 0.0014398577623069286, + "learning_rate": 0.046684162928542286, + "loss": 0.0669, + "num_input_tokens_seen": 38467136, + "step": 29675 + }, + { + "epoch": 1.450199105855911, + "grad_norm": 0.0012009660713374615, + "learning_rate": 0.04664146617926222, + "loss": 0.0864, + "num_input_tokens_seen": 38473088, + "step": 29680 + }, + { + "epoch": 1.450443407519605, + "grad_norm": 0.0020474004559218884, + "learning_rate": 0.046598785369167, + "loss": 0.0736, + "num_input_tokens_seen": 38479488, + "step": 29685 + }, + { + "epoch": 1.4506877091832995, + "grad_norm": 0.0006749614840373397, + "learning_rate": 0.046556120504838434, + "loss": 0.0732, + "num_input_tokens_seen": 38485408, + "step": 29690 + }, + { + "epoch": 1.4509320108469939, + "grad_norm": 0.0008138761622831225, + "learning_rate": 0.04651347159285609, + "loss": 0.0726, + "num_input_tokens_seen": 38491872, + "step": 29695 + }, + { + "epoch": 1.4511763125106882, + "grad_norm": 0.0012180559569969773, + "learning_rate": 0.04647083863979688, + "loss": 0.0697, + "num_input_tokens_seen": 38498240, + "step": 29700 + }, + { + "epoch": 1.4514206141743826, + "grad_norm": 0.0014913418563082814, + "learning_rate": 0.04642822165223538, + "loss": 0.0752, + "num_input_tokens_seen": 38504608, + "step": 29705 + }, + { + "epoch": 1.4516649158380768, + "grad_norm": 0.002043945947661996, + "learning_rate": 0.046385620636743716, + "loss": 0.068, + "num_input_tokens_seen": 38511424, + "step": 29710 + }, + { + "epoch": 1.4519092175017712, + "grad_norm": 0.0019441512413322926, + "learning_rate": 0.04634303559989141, + "loss": 0.0801, + "num_input_tokens_seen": 38517984, + "step": 29715 + }, + { + "epoch": 1.4521535191654655, + "grad_norm": 0.0011486455332487822, + "learning_rate": 0.046300466548245635, + "loss": 0.1214, + "num_input_tokens_seen": 38524736, + "step": 29720 + }, + { + "epoch": 1.45239782082916, + "grad_norm": 0.001974159851670265, + "learning_rate": 0.04625791348837114, + "loss": 0.0885, + "num_input_tokens_seen": 38531104, + "step": 29725 + }, + { + "epoch": 1.452642122492854, + "grad_norm": 0.0010142559185624123, + "learning_rate": 0.046215376426830095, + "loss": 0.0954, + "num_input_tokens_seen": 38538176, + "step": 29730 + }, + { + "epoch": 1.4528864241565485, + "grad_norm": 0.001384071889333427, + "learning_rate": 0.04617285537018219, + "loss": 0.0718, + "num_input_tokens_seen": 38544800, + "step": 29735 + }, + { + "epoch": 1.4531307258202428, + "grad_norm": 0.0011309218825772405, + "learning_rate": 0.046130350324984803, + "loss": 0.0826, + "num_input_tokens_seen": 38551104, + "step": 29740 + }, + { + "epoch": 1.4533750274839372, + "grad_norm": 0.001394416205585003, + "learning_rate": 0.046087861297792666, + "loss": 0.0842, + "num_input_tokens_seen": 38557664, + "step": 29745 + }, + { + "epoch": 1.4536193291476316, + "grad_norm": 0.0015047957422211766, + "learning_rate": 0.0460453882951582, + "loss": 0.0642, + "num_input_tokens_seen": 38564352, + "step": 29750 + }, + { + "epoch": 1.4538636308113257, + "grad_norm": 0.00047319341683760285, + "learning_rate": 0.04600293132363119, + "loss": 0.0693, + "num_input_tokens_seen": 38570688, + "step": 29755 + }, + { + "epoch": 1.4541079324750201, + "grad_norm": 0.000869444222189486, + "learning_rate": 0.045960490389759086, + "loss": 0.0876, + "num_input_tokens_seen": 38577152, + "step": 29760 + }, + { + "epoch": 1.4543522341387145, + "grad_norm": 0.0011641106102615595, + "learning_rate": 0.04591806550008685, + "loss": 0.0782, + "num_input_tokens_seen": 38583520, + "step": 29765 + }, + { + "epoch": 1.4545965358024089, + "grad_norm": 0.0010404819622635841, + "learning_rate": 0.045875656661156825, + "loss": 0.1074, + "num_input_tokens_seen": 38589664, + "step": 29770 + }, + { + "epoch": 1.454840837466103, + "grad_norm": 0.0010679586557671428, + "learning_rate": 0.04583326387950911, + "loss": 0.062, + "num_input_tokens_seen": 38596064, + "step": 29775 + }, + { + "epoch": 1.4550851391297974, + "grad_norm": 0.0010912488214671612, + "learning_rate": 0.0457908871616811, + "loss": 0.0806, + "num_input_tokens_seen": 38602208, + "step": 29780 + }, + { + "epoch": 1.4553294407934918, + "grad_norm": 0.0009829581249505281, + "learning_rate": 0.04574852651420786, + "loss": 0.0873, + "num_input_tokens_seen": 38609024, + "step": 29785 + }, + { + "epoch": 1.4555737424571862, + "grad_norm": 0.0009076777496375144, + "learning_rate": 0.045706181943621985, + "loss": 0.077, + "num_input_tokens_seen": 38615552, + "step": 29790 + }, + { + "epoch": 1.4558180441208806, + "grad_norm": 0.001412243815138936, + "learning_rate": 0.04566385345645344, + "loss": 0.0742, + "num_input_tokens_seen": 38622016, + "step": 29795 + }, + { + "epoch": 1.4560623457845747, + "grad_norm": 0.0008106122259050608, + "learning_rate": 0.04562154105922993, + "loss": 0.0729, + "num_input_tokens_seen": 38629280, + "step": 29800 + }, + { + "epoch": 1.4560623457845747, + "eval_loss": 0.08773259818553925, + "eval_runtime": 403.1738, + "eval_samples_per_second": 90.246, + "eval_steps_per_second": 22.563, + "num_input_tokens_seen": 38629280, + "step": 29800 + }, + { + "epoch": 1.456306647448269, + "grad_norm": 0.0010640136897563934, + "learning_rate": 0.04557924475847642, + "loss": 0.0759, + "num_input_tokens_seen": 38635424, + "step": 29805 + }, + { + "epoch": 1.4565509491119635, + "grad_norm": 0.0011872017057612538, + "learning_rate": 0.04553696456071567, + "loss": 0.0765, + "num_input_tokens_seen": 38642112, + "step": 29810 + }, + { + "epoch": 1.4567952507756579, + "grad_norm": 0.0014303760835900903, + "learning_rate": 0.045494700472467724, + "loss": 0.0841, + "num_input_tokens_seen": 38648736, + "step": 29815 + }, + { + "epoch": 1.457039552439352, + "grad_norm": 0.0007820355240255594, + "learning_rate": 0.04545245250025024, + "loss": 0.0852, + "num_input_tokens_seen": 38655104, + "step": 29820 + }, + { + "epoch": 1.4572838541030464, + "grad_norm": 0.001150690601207316, + "learning_rate": 0.045410220650578384, + "loss": 0.0794, + "num_input_tokens_seen": 38661600, + "step": 29825 + }, + { + "epoch": 1.4575281557667408, + "grad_norm": 0.0024228913243860006, + "learning_rate": 0.04536800492996492, + "loss": 0.0917, + "num_input_tokens_seen": 38668288, + "step": 29830 + }, + { + "epoch": 1.4577724574304352, + "grad_norm": 0.0012293003965169191, + "learning_rate": 0.04532580534491994, + "loss": 0.0776, + "num_input_tokens_seen": 38674560, + "step": 29835 + }, + { + "epoch": 1.4580167590941295, + "grad_norm": 0.0004959730431437492, + "learning_rate": 0.045283621901951183, + "loss": 0.0678, + "num_input_tokens_seen": 38680864, + "step": 29840 + }, + { + "epoch": 1.4582610607578237, + "grad_norm": 0.0009279833757318556, + "learning_rate": 0.04524145460756393, + "loss": 0.0845, + "num_input_tokens_seen": 38687232, + "step": 29845 + }, + { + "epoch": 1.458505362421518, + "grad_norm": 0.000873971323017031, + "learning_rate": 0.045199303468260794, + "loss": 0.0667, + "num_input_tokens_seen": 38693728, + "step": 29850 + }, + { + "epoch": 1.4587496640852124, + "grad_norm": 0.0014577812980860472, + "learning_rate": 0.04515716849054214, + "loss": 0.0889, + "num_input_tokens_seen": 38700256, + "step": 29855 + }, + { + "epoch": 1.4589939657489068, + "grad_norm": 0.0014761548954993486, + "learning_rate": 0.04511504968090558, + "loss": 0.0781, + "num_input_tokens_seen": 38707296, + "step": 29860 + }, + { + "epoch": 1.459238267412601, + "grad_norm": 0.0008834156324155629, + "learning_rate": 0.04507294704584644, + "loss": 0.0754, + "num_input_tokens_seen": 38713696, + "step": 29865 + }, + { + "epoch": 1.4594825690762954, + "grad_norm": 0.001405772753059864, + "learning_rate": 0.04503086059185749, + "loss": 0.0674, + "num_input_tokens_seen": 38720352, + "step": 29870 + }, + { + "epoch": 1.4597268707399897, + "grad_norm": 0.0009643799858167768, + "learning_rate": 0.04498879032542893, + "loss": 0.0773, + "num_input_tokens_seen": 38727136, + "step": 29875 + }, + { + "epoch": 1.4599711724036841, + "grad_norm": 0.0027673363219946623, + "learning_rate": 0.0449467362530486, + "loss": 0.1045, + "num_input_tokens_seen": 38733536, + "step": 29880 + }, + { + "epoch": 1.4602154740673785, + "grad_norm": 0.0013659677933901548, + "learning_rate": 0.04490469838120171, + "loss": 0.0699, + "num_input_tokens_seen": 38740064, + "step": 29885 + }, + { + "epoch": 1.4604597757310727, + "grad_norm": 0.0013575275661423802, + "learning_rate": 0.04486267671637101, + "loss": 0.0792, + "num_input_tokens_seen": 38746304, + "step": 29890 + }, + { + "epoch": 1.460704077394767, + "grad_norm": 0.0012671771692112088, + "learning_rate": 0.04482067126503683, + "loss": 0.0981, + "num_input_tokens_seen": 38752992, + "step": 29895 + }, + { + "epoch": 1.4609483790584614, + "grad_norm": 0.0017992028733715415, + "learning_rate": 0.04477868203367687, + "loss": 0.0936, + "num_input_tokens_seen": 38759520, + "step": 29900 + }, + { + "epoch": 1.4611926807221558, + "grad_norm": 0.0029919343069195747, + "learning_rate": 0.044736709028766426, + "loss": 0.094, + "num_input_tokens_seen": 38766336, + "step": 29905 + }, + { + "epoch": 1.46143698238585, + "grad_norm": 0.0025067850947380066, + "learning_rate": 0.04469475225677832, + "loss": 0.0772, + "num_input_tokens_seen": 38772800, + "step": 29910 + }, + { + "epoch": 1.4616812840495443, + "grad_norm": 0.0009036737610585988, + "learning_rate": 0.04465281172418273, + "loss": 0.095, + "num_input_tokens_seen": 38779648, + "step": 29915 + }, + { + "epoch": 1.4619255857132387, + "grad_norm": 0.001298565766774118, + "learning_rate": 0.044610887437447476, + "loss": 0.0847, + "num_input_tokens_seen": 38786112, + "step": 29920 + }, + { + "epoch": 1.462169887376933, + "grad_norm": 0.0014799418859183788, + "learning_rate": 0.044568979403037744, + "loss": 0.0732, + "num_input_tokens_seen": 38792608, + "step": 29925 + }, + { + "epoch": 1.4624141890406275, + "grad_norm": 0.0015019318088889122, + "learning_rate": 0.04452708762741631, + "loss": 0.1004, + "num_input_tokens_seen": 38799488, + "step": 29930 + }, + { + "epoch": 1.4626584907043216, + "grad_norm": 0.001097405795007944, + "learning_rate": 0.044485212117043475, + "loss": 0.0627, + "num_input_tokens_seen": 38805792, + "step": 29935 + }, + { + "epoch": 1.462902792368016, + "grad_norm": 0.0012197400210425258, + "learning_rate": 0.04444335287837687, + "loss": 0.1148, + "num_input_tokens_seen": 38811744, + "step": 29940 + }, + { + "epoch": 1.4631470940317104, + "grad_norm": 0.0008305628434754908, + "learning_rate": 0.04440150991787179, + "loss": 0.1083, + "num_input_tokens_seen": 38818208, + "step": 29945 + }, + { + "epoch": 1.4633913956954046, + "grad_norm": 0.0011353034060448408, + "learning_rate": 0.04435968324198088, + "loss": 0.0736, + "num_input_tokens_seen": 38824800, + "step": 29950 + }, + { + "epoch": 1.463635697359099, + "grad_norm": 0.0023928158916532993, + "learning_rate": 0.04431787285715442, + "loss": 0.0864, + "num_input_tokens_seen": 38831392, + "step": 29955 + }, + { + "epoch": 1.4638799990227933, + "grad_norm": 0.0018212966388091445, + "learning_rate": 0.04427607876984004, + "loss": 0.0946, + "num_input_tokens_seen": 38837696, + "step": 29960 + }, + { + "epoch": 1.4641243006864877, + "grad_norm": 0.0011102319695055485, + "learning_rate": 0.044234300986482886, + "loss": 0.0714, + "num_input_tokens_seen": 38844256, + "step": 29965 + }, + { + "epoch": 1.464368602350182, + "grad_norm": 0.0008843948016874492, + "learning_rate": 0.04419253951352566, + "loss": 0.0793, + "num_input_tokens_seen": 38850496, + "step": 29970 + }, + { + "epoch": 1.4646129040138764, + "grad_norm": 0.0009567891829647124, + "learning_rate": 0.044150794357408533, + "loss": 0.0801, + "num_input_tokens_seen": 38856352, + "step": 29975 + }, + { + "epoch": 1.4648572056775706, + "grad_norm": 0.0014262922341004014, + "learning_rate": 0.044109065524569065, + "loss": 0.0897, + "num_input_tokens_seen": 38862496, + "step": 29980 + }, + { + "epoch": 1.465101507341265, + "grad_norm": 0.0012162234634160995, + "learning_rate": 0.0440673530214424, + "loss": 0.0897, + "num_input_tokens_seen": 38868512, + "step": 29985 + }, + { + "epoch": 1.4653458090049594, + "grad_norm": 0.0011331361019983888, + "learning_rate": 0.04402565685446117, + "loss": 0.0966, + "num_input_tokens_seen": 38874912, + "step": 29990 + }, + { + "epoch": 1.4655901106686535, + "grad_norm": 0.001358303357847035, + "learning_rate": 0.04398397703005536, + "loss": 0.1062, + "num_input_tokens_seen": 38881344, + "step": 29995 + }, + { + "epoch": 1.465834412332348, + "grad_norm": 0.0017413495806977153, + "learning_rate": 0.043942313554652626, + "loss": 0.0771, + "num_input_tokens_seen": 38887744, + "step": 30000 + }, + { + "epoch": 1.465834412332348, + "eval_loss": 0.08719838410615921, + "eval_runtime": 402.5683, + "eval_samples_per_second": 90.382, + "eval_steps_per_second": 22.597, + "num_input_tokens_seen": 38887744, + "step": 30000 + }, + { + "epoch": 1.4660787139960423, + "grad_norm": 0.000860975356772542, + "learning_rate": 0.0439006664346779, + "loss": 0.0748, + "num_input_tokens_seen": 38893952, + "step": 30005 + }, + { + "epoch": 1.4663230156597367, + "grad_norm": 0.0014302909839898348, + "learning_rate": 0.043859035676553755, + "loss": 0.1077, + "num_input_tokens_seen": 38900064, + "step": 30010 + }, + { + "epoch": 1.466567317323431, + "grad_norm": 0.0008547783363610506, + "learning_rate": 0.043817421286700194, + "loss": 0.1036, + "num_input_tokens_seen": 38906528, + "step": 30015 + }, + { + "epoch": 1.4668116189871254, + "grad_norm": 0.0011126301251351833, + "learning_rate": 0.043775823271534585, + "loss": 0.0613, + "num_input_tokens_seen": 38913248, + "step": 30020 + }, + { + "epoch": 1.4670559206508196, + "grad_norm": 0.0009443470044061542, + "learning_rate": 0.04373424163747197, + "loss": 0.0728, + "num_input_tokens_seen": 38919904, + "step": 30025 + }, + { + "epoch": 1.467300222314514, + "grad_norm": 0.0010627723531797528, + "learning_rate": 0.04369267639092473, + "loss": 0.1139, + "num_input_tokens_seen": 38926240, + "step": 30030 + }, + { + "epoch": 1.4675445239782083, + "grad_norm": 0.0011719027534127235, + "learning_rate": 0.04365112753830268, + "loss": 0.0711, + "num_input_tokens_seen": 38932992, + "step": 30035 + }, + { + "epoch": 1.4677888256419025, + "grad_norm": 0.001359737478196621, + "learning_rate": 0.04360959508601327, + "loss": 0.0857, + "num_input_tokens_seen": 38939328, + "step": 30040 + }, + { + "epoch": 1.4680331273055969, + "grad_norm": 0.0009778633248060942, + "learning_rate": 0.04356807904046123, + "loss": 0.0843, + "num_input_tokens_seen": 38945664, + "step": 30045 + }, + { + "epoch": 1.4682774289692913, + "grad_norm": 0.0009895736584439874, + "learning_rate": 0.04352657940804892, + "loss": 0.0787, + "num_input_tokens_seen": 38951840, + "step": 30050 + }, + { + "epoch": 1.4685217306329856, + "grad_norm": 0.0011931877816095948, + "learning_rate": 0.04348509619517613, + "loss": 0.1018, + "num_input_tokens_seen": 38958400, + "step": 30055 + }, + { + "epoch": 1.46876603229668, + "grad_norm": 0.0007890687556937337, + "learning_rate": 0.04344362940824002, + "loss": 0.082, + "num_input_tokens_seen": 38965120, + "step": 30060 + }, + { + "epoch": 1.4690103339603744, + "grad_norm": 0.001210236456245184, + "learning_rate": 0.04340217905363533, + "loss": 0.0846, + "num_input_tokens_seen": 38971488, + "step": 30065 + }, + { + "epoch": 1.4692546356240686, + "grad_norm": 0.0012781965779140592, + "learning_rate": 0.04336074513775425, + "loss": 0.092, + "num_input_tokens_seen": 38977536, + "step": 30070 + }, + { + "epoch": 1.469498937287763, + "grad_norm": 0.0013901720521971583, + "learning_rate": 0.04331932766698636, + "loss": 0.0685, + "num_input_tokens_seen": 38984096, + "step": 30075 + }, + { + "epoch": 1.4697432389514573, + "grad_norm": 0.0015459171263501048, + "learning_rate": 0.0432779266477188, + "loss": 0.0927, + "num_input_tokens_seen": 38990496, + "step": 30080 + }, + { + "epoch": 1.4699875406151515, + "grad_norm": 0.0008924960275180638, + "learning_rate": 0.04323654208633607, + "loss": 0.0946, + "num_input_tokens_seen": 38997376, + "step": 30085 + }, + { + "epoch": 1.4702318422788458, + "grad_norm": 0.0009236216428689659, + "learning_rate": 0.04319517398922024, + "loss": 0.1091, + "num_input_tokens_seen": 39003968, + "step": 30090 + }, + { + "epoch": 1.4704761439425402, + "grad_norm": 0.0014061733381822705, + "learning_rate": 0.04315382236275079, + "loss": 0.0795, + "num_input_tokens_seen": 39010112, + "step": 30095 + }, + { + "epoch": 1.4707204456062346, + "grad_norm": 0.0014166332548484206, + "learning_rate": 0.043112487213304664, + "loss": 0.1006, + "num_input_tokens_seen": 39016576, + "step": 30100 + }, + { + "epoch": 1.470964747269929, + "grad_norm": 0.0009528872324153781, + "learning_rate": 0.04307116854725618, + "loss": 0.0652, + "num_input_tokens_seen": 39022880, + "step": 30105 + }, + { + "epoch": 1.4712090489336234, + "grad_norm": 0.0019150965381413698, + "learning_rate": 0.043029866370977325, + "loss": 0.0731, + "num_input_tokens_seen": 39029312, + "step": 30110 + }, + { + "epoch": 1.4714533505973175, + "grad_norm": 0.001074412721209228, + "learning_rate": 0.04298858069083728, + "loss": 0.0651, + "num_input_tokens_seen": 39035488, + "step": 30115 + }, + { + "epoch": 1.471697652261012, + "grad_norm": 0.0010034114820882678, + "learning_rate": 0.04294731151320295, + "loss": 0.0686, + "num_input_tokens_seen": 39042304, + "step": 30120 + }, + { + "epoch": 1.4719419539247063, + "grad_norm": 0.0021199346520006657, + "learning_rate": 0.04290605884443841, + "loss": 0.0888, + "num_input_tokens_seen": 39048288, + "step": 30125 + }, + { + "epoch": 1.4721862555884004, + "grad_norm": 0.0007975016487762332, + "learning_rate": 0.04286482269090545, + "loss": 0.064, + "num_input_tokens_seen": 39054912, + "step": 30130 + }, + { + "epoch": 1.4724305572520948, + "grad_norm": 0.0012787265004590154, + "learning_rate": 0.04282360305896323, + "loss": 0.0855, + "num_input_tokens_seen": 39061472, + "step": 30135 + }, + { + "epoch": 1.4726748589157892, + "grad_norm": 0.0009565718355588615, + "learning_rate": 0.04278239995496822, + "loss": 0.1164, + "num_input_tokens_seen": 39067968, + "step": 30140 + }, + { + "epoch": 1.4729191605794836, + "grad_norm": 0.0013547184644266963, + "learning_rate": 0.042741213385274514, + "loss": 0.0813, + "num_input_tokens_seen": 39074400, + "step": 30145 + }, + { + "epoch": 1.473163462243178, + "grad_norm": 0.0012855151435360312, + "learning_rate": 0.04270004335623366, + "loss": 0.1049, + "num_input_tokens_seen": 39080320, + "step": 30150 + }, + { + "epoch": 1.4734077639068723, + "grad_norm": 0.0011306529631838202, + "learning_rate": 0.04265888987419448, + "loss": 0.0935, + "num_input_tokens_seen": 39087456, + "step": 30155 + }, + { + "epoch": 1.4736520655705665, + "grad_norm": 0.0007709242054261267, + "learning_rate": 0.04261775294550346, + "loss": 0.0688, + "num_input_tokens_seen": 39093952, + "step": 30160 + }, + { + "epoch": 1.4738963672342609, + "grad_norm": 0.0012883084127679467, + "learning_rate": 0.042576632576504354, + "loss": 0.0737, + "num_input_tokens_seen": 39100416, + "step": 30165 + }, + { + "epoch": 1.4741406688979553, + "grad_norm": 0.00143773527815938, + "learning_rate": 0.0425355287735385, + "loss": 0.0975, + "num_input_tokens_seen": 39107072, + "step": 30170 + }, + { + "epoch": 1.4743849705616494, + "grad_norm": 0.0014100450789555907, + "learning_rate": 0.0424944415429446, + "loss": 0.0903, + "num_input_tokens_seen": 39113984, + "step": 30175 + }, + { + "epoch": 1.4746292722253438, + "grad_norm": 0.0015030073700472713, + "learning_rate": 0.04245337089105877, + "loss": 0.0831, + "num_input_tokens_seen": 39120480, + "step": 30180 + }, + { + "epoch": 1.4748735738890382, + "grad_norm": 0.0013550296425819397, + "learning_rate": 0.04241231682421467, + "loss": 0.0559, + "num_input_tokens_seen": 39126848, + "step": 30185 + }, + { + "epoch": 1.4751178755527325, + "grad_norm": 0.0010637324303388596, + "learning_rate": 0.04237127934874337, + "loss": 0.0683, + "num_input_tokens_seen": 39133248, + "step": 30190 + }, + { + "epoch": 1.475362177216427, + "grad_norm": 0.0010429422836750746, + "learning_rate": 0.042330258470973305, + "loss": 0.091, + "num_input_tokens_seen": 39140224, + "step": 30195 + }, + { + "epoch": 1.4756064788801213, + "grad_norm": 0.0017082842532545328, + "learning_rate": 0.042289254197230515, + "loss": 0.0929, + "num_input_tokens_seen": 39146016, + "step": 30200 + }, + { + "epoch": 1.4756064788801213, + "eval_loss": 0.08784938603639603, + "eval_runtime": 402.1854, + "eval_samples_per_second": 90.468, + "eval_steps_per_second": 22.619, + "num_input_tokens_seen": 39146016, + "step": 30200 + }, + { + "epoch": 1.4758507805438155, + "grad_norm": 0.001394055550917983, + "learning_rate": 0.04224826653383823, + "loss": 0.0859, + "num_input_tokens_seen": 39152544, + "step": 30205 + }, + { + "epoch": 1.4760950822075098, + "grad_norm": 0.001235123840160668, + "learning_rate": 0.04220729548711735, + "loss": 0.0871, + "num_input_tokens_seen": 39158976, + "step": 30210 + }, + { + "epoch": 1.4763393838712042, + "grad_norm": 0.0016353169921785593, + "learning_rate": 0.04216634106338616, + "loss": 0.1079, + "num_input_tokens_seen": 39165440, + "step": 30215 + }, + { + "epoch": 1.4765836855348984, + "grad_norm": 0.0016508213011547923, + "learning_rate": 0.04212540326896025, + "loss": 0.0797, + "num_input_tokens_seen": 39171872, + "step": 30220 + }, + { + "epoch": 1.4768279871985928, + "grad_norm": 0.0012317444197833538, + "learning_rate": 0.0420844821101528, + "loss": 0.0954, + "num_input_tokens_seen": 39178240, + "step": 30225 + }, + { + "epoch": 1.4770722888622871, + "grad_norm": 0.0012051964877173305, + "learning_rate": 0.04204357759327441, + "loss": 0.0696, + "num_input_tokens_seen": 39184768, + "step": 30230 + }, + { + "epoch": 1.4773165905259815, + "grad_norm": 0.0007868298562243581, + "learning_rate": 0.042002689724632954, + "loss": 0.0848, + "num_input_tokens_seen": 39191136, + "step": 30235 + }, + { + "epoch": 1.477560892189676, + "grad_norm": 0.001545627135783434, + "learning_rate": 0.04196181851053398, + "loss": 0.1107, + "num_input_tokens_seen": 39197312, + "step": 30240 + }, + { + "epoch": 1.47780519385337, + "grad_norm": 0.0014291825937107205, + "learning_rate": 0.041920963957280295, + "loss": 0.089, + "num_input_tokens_seen": 39203712, + "step": 30245 + }, + { + "epoch": 1.4780494955170644, + "grad_norm": 0.0008624944020994008, + "learning_rate": 0.04188012607117212, + "loss": 0.0746, + "num_input_tokens_seen": 39210656, + "step": 30250 + }, + { + "epoch": 1.4782937971807588, + "grad_norm": 0.0026321872137486935, + "learning_rate": 0.04183930485850725, + "loss": 0.0828, + "num_input_tokens_seen": 39217312, + "step": 30255 + }, + { + "epoch": 1.4785380988444532, + "grad_norm": 0.001373200910165906, + "learning_rate": 0.04179850032558078, + "loss": 0.0725, + "num_input_tokens_seen": 39223904, + "step": 30260 + }, + { + "epoch": 1.4787824005081474, + "grad_norm": 0.0013499354245141149, + "learning_rate": 0.041757712478685295, + "loss": 0.0906, + "num_input_tokens_seen": 39230368, + "step": 30265 + }, + { + "epoch": 1.4790267021718417, + "grad_norm": 0.0007618238450959325, + "learning_rate": 0.04171694132411085, + "loss": 0.0739, + "num_input_tokens_seen": 39236832, + "step": 30270 + }, + { + "epoch": 1.4792710038355361, + "grad_norm": 0.0009204060770571232, + "learning_rate": 0.04167618686814479, + "loss": 0.0967, + "num_input_tokens_seen": 39243008, + "step": 30275 + }, + { + "epoch": 1.4795153054992305, + "grad_norm": 0.0010284901363775134, + "learning_rate": 0.041635449117072024, + "loss": 0.0564, + "num_input_tokens_seen": 39249408, + "step": 30280 + }, + { + "epoch": 1.4797596071629249, + "grad_norm": 0.0009718399378471076, + "learning_rate": 0.04159472807717477, + "loss": 0.08, + "num_input_tokens_seen": 39255712, + "step": 30285 + }, + { + "epoch": 1.480003908826619, + "grad_norm": 0.0008127853507176042, + "learning_rate": 0.041554023754732744, + "loss": 0.0694, + "num_input_tokens_seen": 39262656, + "step": 30290 + }, + { + "epoch": 1.4802482104903134, + "grad_norm": 0.0015247021801769733, + "learning_rate": 0.04151333615602311, + "loss": 0.0805, + "num_input_tokens_seen": 39269216, + "step": 30295 + }, + { + "epoch": 1.4804925121540078, + "grad_norm": 0.0016849032836034894, + "learning_rate": 0.04147266528732034, + "loss": 0.0919, + "num_input_tokens_seen": 39275424, + "step": 30300 + }, + { + "epoch": 1.4807368138177022, + "grad_norm": 0.001398916239850223, + "learning_rate": 0.0414320111548964, + "loss": 0.1189, + "num_input_tokens_seen": 39281856, + "step": 30305 + }, + { + "epoch": 1.4809811154813963, + "grad_norm": 0.0015133298002183437, + "learning_rate": 0.04139137376502076, + "loss": 0.0828, + "num_input_tokens_seen": 39288288, + "step": 30310 + }, + { + "epoch": 1.4812254171450907, + "grad_norm": 0.0013996334746479988, + "learning_rate": 0.04135075312396014, + "loss": 0.0861, + "num_input_tokens_seen": 39295424, + "step": 30315 + }, + { + "epoch": 1.481469718808785, + "grad_norm": 0.0012061846209689975, + "learning_rate": 0.04131014923797875, + "loss": 0.0636, + "num_input_tokens_seen": 39301984, + "step": 30320 + }, + { + "epoch": 1.4817140204724795, + "grad_norm": 0.0008223839686252177, + "learning_rate": 0.04126956211333819, + "loss": 0.0661, + "num_input_tokens_seen": 39308576, + "step": 30325 + }, + { + "epoch": 1.4819583221361738, + "grad_norm": 0.001903144526295364, + "learning_rate": 0.041228991756297545, + "loss": 0.0931, + "num_input_tokens_seen": 39314880, + "step": 30330 + }, + { + "epoch": 1.482202623799868, + "grad_norm": 0.0008123563602566719, + "learning_rate": 0.04118843817311332, + "loss": 0.0832, + "num_input_tokens_seen": 39321600, + "step": 30335 + }, + { + "epoch": 1.4824469254635624, + "grad_norm": 0.0013042846694588661, + "learning_rate": 0.0411479013700393, + "loss": 0.0877, + "num_input_tokens_seen": 39328416, + "step": 30340 + }, + { + "epoch": 1.4826912271272568, + "grad_norm": 0.001593786757439375, + "learning_rate": 0.0411073813533268, + "loss": 0.0994, + "num_input_tokens_seen": 39334816, + "step": 30345 + }, + { + "epoch": 1.4829355287909511, + "grad_norm": 0.0012143461499363184, + "learning_rate": 0.04106687812922456, + "loss": 0.0735, + "num_input_tokens_seen": 39341696, + "step": 30350 + }, + { + "epoch": 1.4831798304546453, + "grad_norm": 0.0010176803916692734, + "learning_rate": 0.041026391703978635, + "loss": 0.0871, + "num_input_tokens_seen": 39348384, + "step": 30355 + }, + { + "epoch": 1.4834241321183397, + "grad_norm": 0.0014454069314524531, + "learning_rate": 0.04098592208383259, + "loss": 0.0825, + "num_input_tokens_seen": 39354624, + "step": 30360 + }, + { + "epoch": 1.483668433782034, + "grad_norm": 0.0025879754684865475, + "learning_rate": 0.040945469275027256, + "loss": 0.0964, + "num_input_tokens_seen": 39360928, + "step": 30365 + }, + { + "epoch": 1.4839127354457284, + "grad_norm": 0.0012251246953383088, + "learning_rate": 0.04090503328380104, + "loss": 0.083, + "num_input_tokens_seen": 39367296, + "step": 30370 + }, + { + "epoch": 1.4841570371094228, + "grad_norm": 0.0012763270642608404, + "learning_rate": 0.04086461411638971, + "loss": 0.0735, + "num_input_tokens_seen": 39373888, + "step": 30375 + }, + { + "epoch": 1.484401338773117, + "grad_norm": 0.002294566249474883, + "learning_rate": 0.04082421177902631, + "loss": 0.1051, + "num_input_tokens_seen": 39380512, + "step": 30380 + }, + { + "epoch": 1.4846456404368114, + "grad_norm": 0.001332295942120254, + "learning_rate": 0.04078382627794149, + "loss": 0.0946, + "num_input_tokens_seen": 39386912, + "step": 30385 + }, + { + "epoch": 1.4848899421005057, + "grad_norm": 0.0006400382262654603, + "learning_rate": 0.04074345761936316, + "loss": 0.066, + "num_input_tokens_seen": 39393184, + "step": 30390 + }, + { + "epoch": 1.4851342437642001, + "grad_norm": 0.0012652203440666199, + "learning_rate": 0.04070310580951663, + "loss": 0.0705, + "num_input_tokens_seen": 39399712, + "step": 30395 + }, + { + "epoch": 1.4853785454278943, + "grad_norm": 0.0008004737319424748, + "learning_rate": 0.040662770854624726, + "loss": 0.0642, + "num_input_tokens_seen": 39406240, + "step": 30400 + }, + { + "epoch": 1.4853785454278943, + "eval_loss": 0.08705683052539825, + "eval_runtime": 402.9281, + "eval_samples_per_second": 90.301, + "eval_steps_per_second": 22.577, + "num_input_tokens_seen": 39406240, + "step": 30400 + }, + { + "epoch": 1.4856228470915886, + "grad_norm": 0.0009523361804895103, + "learning_rate": 0.040622452760907535, + "loss": 0.0956, + "num_input_tokens_seen": 39412256, + "step": 30405 + }, + { + "epoch": 1.485867148755283, + "grad_norm": 0.0018916018307209015, + "learning_rate": 0.04058215153458265, + "loss": 0.0793, + "num_input_tokens_seen": 39418592, + "step": 30410 + }, + { + "epoch": 1.4861114504189774, + "grad_norm": 0.001375040621496737, + "learning_rate": 0.04054186718186507, + "loss": 0.0864, + "num_input_tokens_seen": 39425056, + "step": 30415 + }, + { + "epoch": 1.4863557520826718, + "grad_norm": 0.002031746320426464, + "learning_rate": 0.04050159970896708, + "loss": 0.11, + "num_input_tokens_seen": 39431232, + "step": 30420 + }, + { + "epoch": 1.486600053746366, + "grad_norm": 0.0012604667572304606, + "learning_rate": 0.04046134912209843, + "loss": 0.1081, + "num_input_tokens_seen": 39437568, + "step": 30425 + }, + { + "epoch": 1.4868443554100603, + "grad_norm": 0.002456875052303076, + "learning_rate": 0.040421115427466354, + "loss": 0.1209, + "num_input_tokens_seen": 39443648, + "step": 30430 + }, + { + "epoch": 1.4870886570737547, + "grad_norm": 0.001499665784649551, + "learning_rate": 0.04038089863127529, + "loss": 0.0917, + "num_input_tokens_seen": 39450496, + "step": 30435 + }, + { + "epoch": 1.487332958737449, + "grad_norm": 0.001799926394596696, + "learning_rate": 0.04034069873972727, + "loss": 0.1045, + "num_input_tokens_seen": 39457408, + "step": 30440 + }, + { + "epoch": 1.4875772604011432, + "grad_norm": 0.0010672216303646564, + "learning_rate": 0.040300515759021514, + "loss": 0.0862, + "num_input_tokens_seen": 39463712, + "step": 30445 + }, + { + "epoch": 1.4878215620648376, + "grad_norm": 0.0013875728473067284, + "learning_rate": 0.04026034969535478, + "loss": 0.0701, + "num_input_tokens_seen": 39470368, + "step": 30450 + }, + { + "epoch": 1.488065863728532, + "grad_norm": 0.0011007522698491812, + "learning_rate": 0.040220200554921266, + "loss": 0.0785, + "num_input_tokens_seen": 39477184, + "step": 30455 + }, + { + "epoch": 1.4883101653922264, + "grad_norm": 0.0009358122479170561, + "learning_rate": 0.0401800683439124, + "loss": 0.0941, + "num_input_tokens_seen": 39483424, + "step": 30460 + }, + { + "epoch": 1.4885544670559208, + "grad_norm": 0.0017209749203175306, + "learning_rate": 0.04013995306851704, + "loss": 0.0871, + "num_input_tokens_seen": 39490368, + "step": 30465 + }, + { + "epoch": 1.488798768719615, + "grad_norm": 0.0007264286396093667, + "learning_rate": 0.040099854734921545, + "loss": 0.0559, + "num_input_tokens_seen": 39497056, + "step": 30470 + }, + { + "epoch": 1.4890430703833093, + "grad_norm": 0.0011127465404570103, + "learning_rate": 0.0400597733493095, + "loss": 0.0965, + "num_input_tokens_seen": 39503072, + "step": 30475 + }, + { + "epoch": 1.4892873720470037, + "grad_norm": 0.001151003409177065, + "learning_rate": 0.04001970891786203, + "loss": 0.0699, + "num_input_tokens_seen": 39509856, + "step": 30480 + }, + { + "epoch": 1.4895316737106978, + "grad_norm": 0.0016762195155024529, + "learning_rate": 0.03997966144675752, + "loss": 0.0671, + "num_input_tokens_seen": 39516544, + "step": 30485 + }, + { + "epoch": 1.4897759753743922, + "grad_norm": 0.0014914795756340027, + "learning_rate": 0.039939630942171796, + "loss": 0.0884, + "num_input_tokens_seen": 39523200, + "step": 30490 + }, + { + "epoch": 1.4900202770380866, + "grad_norm": 0.0009818655671551824, + "learning_rate": 0.03989961741027815, + "loss": 0.0615, + "num_input_tokens_seen": 39529728, + "step": 30495 + }, + { + "epoch": 1.490264578701781, + "grad_norm": 0.0014830502914264798, + "learning_rate": 0.03985962085724704, + "loss": 0.0955, + "num_input_tokens_seen": 39536256, + "step": 30500 + }, + { + "epoch": 1.4905088803654754, + "grad_norm": 0.0018147911177948117, + "learning_rate": 0.03981964128924656, + "loss": 0.0992, + "num_input_tokens_seen": 39542368, + "step": 30505 + }, + { + "epoch": 1.4907531820291697, + "grad_norm": 0.0012414460070431232, + "learning_rate": 0.03977967871244197, + "loss": 0.0608, + "num_input_tokens_seen": 39548960, + "step": 30510 + }, + { + "epoch": 1.490997483692864, + "grad_norm": 0.0008633933612145483, + "learning_rate": 0.03973973313299602, + "loss": 0.0728, + "num_input_tokens_seen": 39555328, + "step": 30515 + }, + { + "epoch": 1.4912417853565583, + "grad_norm": 0.002361691789701581, + "learning_rate": 0.0396998045570689, + "loss": 0.1344, + "num_input_tokens_seen": 39561760, + "step": 30520 + }, + { + "epoch": 1.4914860870202526, + "grad_norm": 0.002626015804708004, + "learning_rate": 0.03965989299081798, + "loss": 0.0862, + "num_input_tokens_seen": 39568352, + "step": 30525 + }, + { + "epoch": 1.4917303886839468, + "grad_norm": 0.0012915765400975943, + "learning_rate": 0.039619998440398235, + "loss": 0.1015, + "num_input_tokens_seen": 39574496, + "step": 30530 + }, + { + "epoch": 1.4919746903476412, + "grad_norm": 0.0010782756144180894, + "learning_rate": 0.03958012091196184, + "loss": 0.0731, + "num_input_tokens_seen": 39581152, + "step": 30535 + }, + { + "epoch": 1.4922189920113356, + "grad_norm": 0.0012324067065492272, + "learning_rate": 0.039540260411658396, + "loss": 0.083, + "num_input_tokens_seen": 39587552, + "step": 30540 + }, + { + "epoch": 1.49246329367503, + "grad_norm": 0.0007207982707768679, + "learning_rate": 0.03950041694563496, + "loss": 0.0719, + "num_input_tokens_seen": 39593984, + "step": 30545 + }, + { + "epoch": 1.4927075953387243, + "grad_norm": 0.0012383671710267663, + "learning_rate": 0.0394605905200358, + "loss": 0.0761, + "num_input_tokens_seen": 39600160, + "step": 30550 + }, + { + "epoch": 1.4929518970024187, + "grad_norm": 0.0009193563018925488, + "learning_rate": 0.03942078114100272, + "loss": 0.0743, + "num_input_tokens_seen": 39606528, + "step": 30555 + }, + { + "epoch": 1.4931961986661129, + "grad_norm": 0.0030333076138049364, + "learning_rate": 0.03938098881467485, + "loss": 0.0751, + "num_input_tokens_seen": 39613088, + "step": 30560 + }, + { + "epoch": 1.4934405003298072, + "grad_norm": 0.0007567111751995981, + "learning_rate": 0.039341213547188586, + "loss": 0.0571, + "num_input_tokens_seen": 39620128, + "step": 30565 + }, + { + "epoch": 1.4936848019935016, + "grad_norm": 0.0008380998042412102, + "learning_rate": 0.03930145534467782, + "loss": 0.0678, + "num_input_tokens_seen": 39626624, + "step": 30570 + }, + { + "epoch": 1.4939291036571958, + "grad_norm": 0.00201870477758348, + "learning_rate": 0.0392617142132738, + "loss": 0.0898, + "num_input_tokens_seen": 39632704, + "step": 30575 + }, + { + "epoch": 1.4941734053208902, + "grad_norm": 0.0006166244274936616, + "learning_rate": 0.03922199015910504, + "loss": 0.0641, + "num_input_tokens_seen": 39638720, + "step": 30580 + }, + { + "epoch": 1.4944177069845845, + "grad_norm": 0.0015325377462431788, + "learning_rate": 0.039182283188297556, + "loss": 0.0777, + "num_input_tokens_seen": 39645440, + "step": 30585 + }, + { + "epoch": 1.494662008648279, + "grad_norm": 0.001002576551400125, + "learning_rate": 0.039142593306974595, + "loss": 0.0986, + "num_input_tokens_seen": 39652128, + "step": 30590 + }, + { + "epoch": 1.4949063103119733, + "grad_norm": 0.002101271180436015, + "learning_rate": 0.039102920521256856, + "loss": 0.0715, + "num_input_tokens_seen": 39658336, + "step": 30595 + }, + { + "epoch": 1.4951506119756677, + "grad_norm": 0.0019247295567765832, + "learning_rate": 0.03906326483726243, + "loss": 0.0639, + "num_input_tokens_seen": 39664736, + "step": 30600 + }, + { + "epoch": 1.4951506119756677, + "eval_loss": 0.08708000928163528, + "eval_runtime": 402.7212, + "eval_samples_per_second": 90.348, + "eval_steps_per_second": 22.589, + "num_input_tokens_seen": 39664736, + "step": 30600 + }, + { + "epoch": 1.4953949136393618, + "grad_norm": 0.0009488352225162089, + "learning_rate": 0.039023626261106704, + "loss": 0.0794, + "num_input_tokens_seen": 39671520, + "step": 30605 + }, + { + "epoch": 1.4956392153030562, + "grad_norm": 0.0007170360186137259, + "learning_rate": 0.03898400479890237, + "loss": 0.0955, + "num_input_tokens_seen": 39678016, + "step": 30610 + }, + { + "epoch": 1.4958835169667506, + "grad_norm": 0.001961632864549756, + "learning_rate": 0.038944400456759655, + "loss": 0.1082, + "num_input_tokens_seen": 39684512, + "step": 30615 + }, + { + "epoch": 1.4961278186304448, + "grad_norm": 0.0009080914896912873, + "learning_rate": 0.038904813240785964, + "loss": 0.0704, + "num_input_tokens_seen": 39690912, + "step": 30620 + }, + { + "epoch": 1.4963721202941391, + "grad_norm": 0.001242755795828998, + "learning_rate": 0.03886524315708621, + "loss": 0.0678, + "num_input_tokens_seen": 39697568, + "step": 30625 + }, + { + "epoch": 1.4966164219578335, + "grad_norm": 0.0014869585866108537, + "learning_rate": 0.03882569021176255, + "loss": 0.0834, + "num_input_tokens_seen": 39704032, + "step": 30630 + }, + { + "epoch": 1.4968607236215279, + "grad_norm": 0.001722906599752605, + "learning_rate": 0.038786154410914535, + "loss": 0.0696, + "num_input_tokens_seen": 39710336, + "step": 30635 + }, + { + "epoch": 1.4971050252852223, + "grad_norm": 0.001308867591433227, + "learning_rate": 0.03874663576063917, + "loss": 0.0785, + "num_input_tokens_seen": 39716736, + "step": 30640 + }, + { + "epoch": 1.4973493269489166, + "grad_norm": 0.0017692578257992864, + "learning_rate": 0.038707134267030624, + "loss": 0.0649, + "num_input_tokens_seen": 39723648, + "step": 30645 + }, + { + "epoch": 1.4975936286126108, + "grad_norm": 0.0012001435970887542, + "learning_rate": 0.038667649936180555, + "loss": 0.0847, + "num_input_tokens_seen": 39730240, + "step": 30650 + }, + { + "epoch": 1.4978379302763052, + "grad_norm": 0.0016278987750411034, + "learning_rate": 0.038628182774178, + "loss": 0.0856, + "num_input_tokens_seen": 39736480, + "step": 30655 + }, + { + "epoch": 1.4980822319399996, + "grad_norm": 0.001532418536953628, + "learning_rate": 0.038588732787109226, + "loss": 0.0934, + "num_input_tokens_seen": 39742848, + "step": 30660 + }, + { + "epoch": 1.4983265336036937, + "grad_norm": 0.0012047896161675453, + "learning_rate": 0.03854929998105795, + "loss": 0.0729, + "num_input_tokens_seen": 39749280, + "step": 30665 + }, + { + "epoch": 1.498570835267388, + "grad_norm": 0.0006294584600254893, + "learning_rate": 0.03850988436210518, + "loss": 0.048, + "num_input_tokens_seen": 39755808, + "step": 30670 + }, + { + "epoch": 1.4988151369310825, + "grad_norm": 0.0018565405625849962, + "learning_rate": 0.03847048593632933, + "loss": 0.0863, + "num_input_tokens_seen": 39761920, + "step": 30675 + }, + { + "epoch": 1.4990594385947769, + "grad_norm": 0.001359442831017077, + "learning_rate": 0.038431104709806096, + "loss": 0.098, + "num_input_tokens_seen": 39768512, + "step": 30680 + }, + { + "epoch": 1.4993037402584712, + "grad_norm": 0.001193055184558034, + "learning_rate": 0.0383917406886086, + "loss": 0.0931, + "num_input_tokens_seen": 39774464, + "step": 30685 + }, + { + "epoch": 1.4995480419221656, + "grad_norm": 0.0018519886070862412, + "learning_rate": 0.03835239387880722, + "loss": 0.1045, + "num_input_tokens_seen": 39780704, + "step": 30690 + }, + { + "epoch": 1.4997923435858598, + "grad_norm": 0.0009455799008719623, + "learning_rate": 0.03831306428646979, + "loss": 0.116, + "num_input_tokens_seen": 39787328, + "step": 30695 + }, + { + "epoch": 1.5000366452495542, + "grad_norm": 0.0010733954841271043, + "learning_rate": 0.03827375191766135, + "loss": 0.078, + "num_input_tokens_seen": 39793792, + "step": 30700 + }, + { + "epoch": 1.5002809469132483, + "grad_norm": 0.0012871656799688935, + "learning_rate": 0.03823445677844446, + "loss": 0.1061, + "num_input_tokens_seen": 39799968, + "step": 30705 + }, + { + "epoch": 1.5005252485769427, + "grad_norm": 0.0022239182144403458, + "learning_rate": 0.03819517887487881, + "loss": 0.0981, + "num_input_tokens_seen": 39806496, + "step": 30710 + }, + { + "epoch": 1.500769550240637, + "grad_norm": 0.0014834636822342873, + "learning_rate": 0.03815591821302161, + "loss": 0.0862, + "num_input_tokens_seen": 39812704, + "step": 30715 + }, + { + "epoch": 1.5010138519043315, + "grad_norm": 0.0009011050569824874, + "learning_rate": 0.03811667479892739, + "loss": 0.066, + "num_input_tokens_seen": 39819136, + "step": 30720 + }, + { + "epoch": 1.5012581535680258, + "grad_norm": 0.0026168155018240213, + "learning_rate": 0.03807744863864788, + "loss": 0.1071, + "num_input_tokens_seen": 39826272, + "step": 30725 + }, + { + "epoch": 1.5015024552317202, + "grad_norm": 0.0017069298774003983, + "learning_rate": 0.03803823973823229, + "loss": 0.1136, + "num_input_tokens_seen": 39832672, + "step": 30730 + }, + { + "epoch": 1.5017467568954146, + "grad_norm": 0.001112868427298963, + "learning_rate": 0.03799904810372719, + "loss": 0.0668, + "num_input_tokens_seen": 39839456, + "step": 30735 + }, + { + "epoch": 1.5019910585591087, + "grad_norm": 0.0013022735947743058, + "learning_rate": 0.03795987374117632, + "loss": 0.0592, + "num_input_tokens_seen": 39845600, + "step": 30740 + }, + { + "epoch": 1.5022353602228031, + "grad_norm": 0.0014920339453965425, + "learning_rate": 0.03792071665662093, + "loss": 0.0799, + "num_input_tokens_seen": 39851808, + "step": 30745 + }, + { + "epoch": 1.5024796618864973, + "grad_norm": 0.0007835807045921683, + "learning_rate": 0.03788157685609952, + "loss": 0.076, + "num_input_tokens_seen": 39857888, + "step": 30750 + }, + { + "epoch": 1.5027239635501917, + "grad_norm": 0.0007453282014466822, + "learning_rate": 0.037842454345647876, + "loss": 0.1034, + "num_input_tokens_seen": 39864288, + "step": 30755 + }, + { + "epoch": 1.502968265213886, + "grad_norm": 0.001430520904250443, + "learning_rate": 0.03780334913129929, + "loss": 0.0887, + "num_input_tokens_seen": 39870336, + "step": 30760 + }, + { + "epoch": 1.5032125668775804, + "grad_norm": 0.0012331637553870678, + "learning_rate": 0.037764261219084175, + "loss": 0.0945, + "num_input_tokens_seen": 39876672, + "step": 30765 + }, + { + "epoch": 1.5034568685412748, + "grad_norm": 0.0013751628575846553, + "learning_rate": 0.037725190615030414, + "loss": 0.0792, + "num_input_tokens_seen": 39883168, + "step": 30770 + }, + { + "epoch": 1.5037011702049692, + "grad_norm": 0.0013316039694473147, + "learning_rate": 0.037686137325163224, + "loss": 0.0662, + "num_input_tokens_seen": 39889952, + "step": 30775 + }, + { + "epoch": 1.5039454718686636, + "grad_norm": 0.0014069508761167526, + "learning_rate": 0.037647101355505065, + "loss": 0.0945, + "num_input_tokens_seen": 39896416, + "step": 30780 + }, + { + "epoch": 1.5041897735323577, + "grad_norm": 0.000899620761629194, + "learning_rate": 0.03760808271207581, + "loss": 0.0686, + "num_input_tokens_seen": 39902624, + "step": 30785 + }, + { + "epoch": 1.504434075196052, + "grad_norm": 0.00096130307065323, + "learning_rate": 0.03756908140089258, + "loss": 0.0614, + "num_input_tokens_seen": 39909408, + "step": 30790 + }, + { + "epoch": 1.5046783768597463, + "grad_norm": 0.000945123378187418, + "learning_rate": 0.03753009742796989, + "loss": 0.0753, + "num_input_tokens_seen": 39915584, + "step": 30795 + }, + { + "epoch": 1.5049226785234406, + "grad_norm": 0.00115196465048939, + "learning_rate": 0.037491130799319615, + "loss": 0.0726, + "num_input_tokens_seen": 39922240, + "step": 30800 + }, + { + "epoch": 1.5049226785234406, + "eval_loss": 0.08693520724773407, + "eval_runtime": 402.2642, + "eval_samples_per_second": 90.451, + "eval_steps_per_second": 22.614, + "num_input_tokens_seen": 39922240, + "step": 30800 + }, + { + "epoch": 1.505166980187135, + "grad_norm": 0.0010042678331956267, + "learning_rate": 0.03745218152095079, + "loss": 0.0805, + "num_input_tokens_seen": 39928864, + "step": 30805 + }, + { + "epoch": 1.5054112818508294, + "grad_norm": 0.0015617342432960868, + "learning_rate": 0.037413249598869935, + "loss": 0.0676, + "num_input_tokens_seen": 39935584, + "step": 30810 + }, + { + "epoch": 1.5056555835145238, + "grad_norm": 0.0018322734395042062, + "learning_rate": 0.037374335039080886, + "loss": 0.1175, + "num_input_tokens_seen": 39942304, + "step": 30815 + }, + { + "epoch": 1.5058998851782182, + "grad_norm": 0.0011079130927100778, + "learning_rate": 0.037335437847584724, + "loss": 0.0619, + "num_input_tokens_seen": 39949152, + "step": 30820 + }, + { + "epoch": 1.5061441868419125, + "grad_norm": 0.001605542958714068, + "learning_rate": 0.03729655803037983, + "loss": 0.0969, + "num_input_tokens_seen": 39955616, + "step": 30825 + }, + { + "epoch": 1.5063884885056067, + "grad_norm": 0.003004588419571519, + "learning_rate": 0.03725769559346207, + "loss": 0.0894, + "num_input_tokens_seen": 39962144, + "step": 30830 + }, + { + "epoch": 1.506632790169301, + "grad_norm": 0.001199806109070778, + "learning_rate": 0.03721885054282439, + "loss": 0.0637, + "num_input_tokens_seen": 39969024, + "step": 30835 + }, + { + "epoch": 1.5068770918329952, + "grad_norm": 0.002115659648552537, + "learning_rate": 0.03718002288445731, + "loss": 0.0871, + "num_input_tokens_seen": 39975648, + "step": 30840 + }, + { + "epoch": 1.5071213934966896, + "grad_norm": 0.0006289199809543788, + "learning_rate": 0.03714121262434844, + "loss": 0.0895, + "num_input_tokens_seen": 39981888, + "step": 30845 + }, + { + "epoch": 1.507365695160384, + "grad_norm": 0.0007012219866737723, + "learning_rate": 0.037102419768482844, + "loss": 0.087, + "num_input_tokens_seen": 39988320, + "step": 30850 + }, + { + "epoch": 1.5076099968240784, + "grad_norm": 0.002098178956657648, + "learning_rate": 0.03706364432284293, + "loss": 0.0971, + "num_input_tokens_seen": 39994624, + "step": 30855 + }, + { + "epoch": 1.5078542984877727, + "grad_norm": 0.0009346086881123483, + "learning_rate": 0.03702488629340828, + "loss": 0.0867, + "num_input_tokens_seen": 40000928, + "step": 30860 + }, + { + "epoch": 1.5080986001514671, + "grad_norm": 0.0019240471301600337, + "learning_rate": 0.036986145686155915, + "loss": 0.1052, + "num_input_tokens_seen": 40006944, + "step": 30865 + }, + { + "epoch": 1.5083429018151615, + "grad_norm": 0.0010816697031259537, + "learning_rate": 0.036947422507060075, + "loss": 0.0746, + "num_input_tokens_seen": 40013184, + "step": 30870 + }, + { + "epoch": 1.5085872034788557, + "grad_norm": 0.0013420362956821918, + "learning_rate": 0.0369087167620924, + "loss": 0.0925, + "num_input_tokens_seen": 40019808, + "step": 30875 + }, + { + "epoch": 1.50883150514255, + "grad_norm": 0.0018077305285260081, + "learning_rate": 0.03687002845722183, + "loss": 0.0997, + "num_input_tokens_seen": 40026560, + "step": 30880 + }, + { + "epoch": 1.5090758068062442, + "grad_norm": 0.001691638259217143, + "learning_rate": 0.03683135759841451, + "loss": 0.0813, + "num_input_tokens_seen": 40033024, + "step": 30885 + }, + { + "epoch": 1.5093201084699386, + "grad_norm": 0.0015851381467655301, + "learning_rate": 0.03679270419163406, + "loss": 0.0754, + "num_input_tokens_seen": 40039520, + "step": 30890 + }, + { + "epoch": 1.509564410133633, + "grad_norm": 0.0017961778212338686, + "learning_rate": 0.03675406824284127, + "loss": 0.0928, + "num_input_tokens_seen": 40045568, + "step": 30895 + }, + { + "epoch": 1.5098087117973273, + "grad_norm": 0.0010057301260530949, + "learning_rate": 0.03671544975799425, + "loss": 0.0925, + "num_input_tokens_seen": 40051648, + "step": 30900 + }, + { + "epoch": 1.5100530134610217, + "grad_norm": 0.001045412034727633, + "learning_rate": 0.03667684874304854, + "loss": 0.0765, + "num_input_tokens_seen": 40058080, + "step": 30905 + }, + { + "epoch": 1.510297315124716, + "grad_norm": 0.0009362325654365122, + "learning_rate": 0.03663826520395683, + "loss": 0.0859, + "num_input_tokens_seen": 40064096, + "step": 30910 + }, + { + "epoch": 1.5105416167884105, + "grad_norm": 0.001102737500332296, + "learning_rate": 0.03659969914666922, + "loss": 0.0553, + "num_input_tokens_seen": 40070592, + "step": 30915 + }, + { + "epoch": 1.5107859184521046, + "grad_norm": 0.0014338254695758224, + "learning_rate": 0.036561150577133106, + "loss": 0.1009, + "num_input_tokens_seen": 40076672, + "step": 30920 + }, + { + "epoch": 1.511030220115799, + "grad_norm": 0.0012178861070424318, + "learning_rate": 0.036522619501293103, + "loss": 0.0767, + "num_input_tokens_seen": 40083424, + "step": 30925 + }, + { + "epoch": 1.5112745217794932, + "grad_norm": 0.0010413900017738342, + "learning_rate": 0.03648410592509122, + "loss": 0.0742, + "num_input_tokens_seen": 40089728, + "step": 30930 + }, + { + "epoch": 1.5115188234431876, + "grad_norm": 0.0013622365659102798, + "learning_rate": 0.03644560985446676, + "loss": 0.0853, + "num_input_tokens_seen": 40096768, + "step": 30935 + }, + { + "epoch": 1.511763125106882, + "grad_norm": 0.0013868537498638034, + "learning_rate": 0.036407131295356256, + "loss": 0.0673, + "num_input_tokens_seen": 40103648, + "step": 30940 + }, + { + "epoch": 1.5120074267705763, + "grad_norm": 0.0018356924410909414, + "learning_rate": 0.03636867025369362, + "loss": 0.0671, + "num_input_tokens_seen": 40110144, + "step": 30945 + }, + { + "epoch": 1.5122517284342707, + "grad_norm": 0.0011341541539877653, + "learning_rate": 0.03633022673540999, + "loss": 0.0777, + "num_input_tokens_seen": 40116160, + "step": 30950 + }, + { + "epoch": 1.512496030097965, + "grad_norm": 0.0015568218659609556, + "learning_rate": 0.03629180074643385, + "loss": 0.0776, + "num_input_tokens_seen": 40122784, + "step": 30955 + }, + { + "epoch": 1.5127403317616595, + "grad_norm": 0.0008656425634399056, + "learning_rate": 0.03625339229269102, + "loss": 0.0577, + "num_input_tokens_seen": 40129504, + "step": 30960 + }, + { + "epoch": 1.5129846334253536, + "grad_norm": 0.0011121139395982027, + "learning_rate": 0.036215001380104535, + "loss": 0.0856, + "num_input_tokens_seen": 40136352, + "step": 30965 + }, + { + "epoch": 1.513228935089048, + "grad_norm": 0.0007484292727895081, + "learning_rate": 0.03617662801459471, + "loss": 0.0754, + "num_input_tokens_seen": 40142848, + "step": 30970 + }, + { + "epoch": 1.5134732367527421, + "grad_norm": 0.0009054560796357691, + "learning_rate": 0.036138272202079276, + "loss": 0.0855, + "num_input_tokens_seen": 40149376, + "step": 30975 + }, + { + "epoch": 1.5137175384164365, + "grad_norm": 0.0017505211289972067, + "learning_rate": 0.036099933948473106, + "loss": 0.0825, + "num_input_tokens_seen": 40155872, + "step": 30980 + }, + { + "epoch": 1.513961840080131, + "grad_norm": 0.0010949871502816677, + "learning_rate": 0.03606161325968851, + "loss": 0.0799, + "num_input_tokens_seen": 40162144, + "step": 30985 + }, + { + "epoch": 1.5142061417438253, + "grad_norm": 0.001668871147558093, + "learning_rate": 0.03602331014163496, + "loss": 0.0714, + "num_input_tokens_seen": 40168800, + "step": 30990 + }, + { + "epoch": 1.5144504434075197, + "grad_norm": 0.0016826344653964043, + "learning_rate": 0.035985024600219295, + "loss": 0.0768, + "num_input_tokens_seen": 40175264, + "step": 30995 + }, + { + "epoch": 1.514694745071214, + "grad_norm": 0.0008401534869335592, + "learning_rate": 0.03594675664134569, + "loss": 0.0909, + "num_input_tokens_seen": 40181504, + "step": 31000 + }, + { + "epoch": 1.514694745071214, + "eval_loss": 0.08594069629907608, + "eval_runtime": 402.3246, + "eval_samples_per_second": 90.437, + "eval_steps_per_second": 22.611, + "num_input_tokens_seen": 40181504, + "step": 31000 + }, + { + "epoch": 1.5149390467349084, + "grad_norm": 0.0008410470909439027, + "learning_rate": 0.03590850627091545, + "loss": 0.0855, + "num_input_tokens_seen": 40187968, + "step": 31005 + }, + { + "epoch": 1.5151833483986026, + "grad_norm": 0.001108837779611349, + "learning_rate": 0.03587027349482731, + "loss": 0.1058, + "num_input_tokens_seen": 40195136, + "step": 31010 + }, + { + "epoch": 1.515427650062297, + "grad_norm": 0.0010861021000891924, + "learning_rate": 0.035832058318977275, + "loss": 0.0812, + "num_input_tokens_seen": 40201312, + "step": 31015 + }, + { + "epoch": 1.5156719517259911, + "grad_norm": 0.0011724577052518725, + "learning_rate": 0.03579386074925853, + "loss": 0.0908, + "num_input_tokens_seen": 40207520, + "step": 31020 + }, + { + "epoch": 1.5159162533896855, + "grad_norm": 0.0011912262998521328, + "learning_rate": 0.035755680791561696, + "loss": 0.0977, + "num_input_tokens_seen": 40214016, + "step": 31025 + }, + { + "epoch": 1.5161605550533799, + "grad_norm": 0.0005656147841364145, + "learning_rate": 0.03571751845177454, + "loss": 0.0655, + "num_input_tokens_seen": 40220352, + "step": 31030 + }, + { + "epoch": 1.5164048567170743, + "grad_norm": 0.001004135818220675, + "learning_rate": 0.03567937373578225, + "loss": 0.0875, + "num_input_tokens_seen": 40226976, + "step": 31035 + }, + { + "epoch": 1.5166491583807686, + "grad_norm": 0.0014074642676860094, + "learning_rate": 0.03564124664946711, + "loss": 0.0817, + "num_input_tokens_seen": 40233952, + "step": 31040 + }, + { + "epoch": 1.516893460044463, + "grad_norm": 0.0015981984324753284, + "learning_rate": 0.035603137198708924, + "loss": 0.0708, + "num_input_tokens_seen": 40240736, + "step": 31045 + }, + { + "epoch": 1.5171377617081574, + "grad_norm": 0.001112818019464612, + "learning_rate": 0.035565045389384514, + "loss": 0.1165, + "num_input_tokens_seen": 40247232, + "step": 31050 + }, + { + "epoch": 1.5173820633718516, + "grad_norm": 0.0012580535840243101, + "learning_rate": 0.03552697122736823, + "loss": 0.0764, + "num_input_tokens_seen": 40253920, + "step": 31055 + }, + { + "epoch": 1.517626365035546, + "grad_norm": 0.001554334769025445, + "learning_rate": 0.03548891471853153, + "loss": 0.0951, + "num_input_tokens_seen": 40260064, + "step": 31060 + }, + { + "epoch": 1.51787066669924, + "grad_norm": 0.0012639887863770127, + "learning_rate": 0.03545087586874322, + "loss": 0.071, + "num_input_tokens_seen": 40266464, + "step": 31065 + }, + { + "epoch": 1.5181149683629345, + "grad_norm": 0.0015302980318665504, + "learning_rate": 0.03541285468386935, + "loss": 0.097, + "num_input_tokens_seen": 40272928, + "step": 31070 + }, + { + "epoch": 1.5183592700266288, + "grad_norm": 0.0009121674229390919, + "learning_rate": 0.03537485116977327, + "loss": 0.0776, + "num_input_tokens_seen": 40279136, + "step": 31075 + }, + { + "epoch": 1.5186035716903232, + "grad_norm": 0.0013629380846396089, + "learning_rate": 0.03533686533231565, + "loss": 0.1016, + "num_input_tokens_seen": 40285760, + "step": 31080 + }, + { + "epoch": 1.5188478733540176, + "grad_norm": 0.001183652668260038, + "learning_rate": 0.0352988971773543, + "loss": 0.1294, + "num_input_tokens_seen": 40291904, + "step": 31085 + }, + { + "epoch": 1.519092175017712, + "grad_norm": 0.0035529809538275003, + "learning_rate": 0.03526094671074443, + "loss": 0.0806, + "num_input_tokens_seen": 40298272, + "step": 31090 + }, + { + "epoch": 1.5193364766814061, + "grad_norm": 0.0010364663321524858, + "learning_rate": 0.03522301393833852, + "loss": 0.0854, + "num_input_tokens_seen": 40304576, + "step": 31095 + }, + { + "epoch": 1.5195807783451005, + "grad_norm": 0.0015204488299787045, + "learning_rate": 0.035185098865986204, + "loss": 0.0883, + "num_input_tokens_seen": 40310944, + "step": 31100 + }, + { + "epoch": 1.519825080008795, + "grad_norm": 0.001029456965625286, + "learning_rate": 0.03514720149953453, + "loss": 0.0821, + "num_input_tokens_seen": 40317152, + "step": 31105 + }, + { + "epoch": 1.520069381672489, + "grad_norm": 0.0008465711725875735, + "learning_rate": 0.03510932184482773, + "loss": 0.0866, + "num_input_tokens_seen": 40323296, + "step": 31110 + }, + { + "epoch": 1.5203136833361834, + "grad_norm": 0.0007476642495021224, + "learning_rate": 0.03507145990770724, + "loss": 0.1006, + "num_input_tokens_seen": 40329792, + "step": 31115 + }, + { + "epoch": 1.5205579849998778, + "grad_norm": 0.0009389373590238392, + "learning_rate": 0.035033615694011984, + "loss": 0.0805, + "num_input_tokens_seen": 40336288, + "step": 31120 + }, + { + "epoch": 1.5208022866635722, + "grad_norm": 0.001865740050561726, + "learning_rate": 0.03499578920957788, + "loss": 0.0824, + "num_input_tokens_seen": 40342880, + "step": 31125 + }, + { + "epoch": 1.5210465883272666, + "grad_norm": 0.0011327676475048065, + "learning_rate": 0.034957980460238375, + "loss": 0.0822, + "num_input_tokens_seen": 40349152, + "step": 31130 + }, + { + "epoch": 1.521290889990961, + "grad_norm": 0.0011177904671058059, + "learning_rate": 0.03492018945182393, + "loss": 0.0766, + "num_input_tokens_seen": 40355424, + "step": 31135 + }, + { + "epoch": 1.5215351916546551, + "grad_norm": 0.0008354930905625224, + "learning_rate": 0.03488241619016247, + "loss": 0.0866, + "num_input_tokens_seen": 40362304, + "step": 31140 + }, + { + "epoch": 1.5217794933183495, + "grad_norm": 0.0015529717784374952, + "learning_rate": 0.03484466068107913, + "loss": 0.0793, + "num_input_tokens_seen": 40369088, + "step": 31145 + }, + { + "epoch": 1.5220237949820439, + "grad_norm": 0.0007790831732563674, + "learning_rate": 0.034806922930396195, + "loss": 0.0577, + "num_input_tokens_seen": 40375808, + "step": 31150 + }, + { + "epoch": 1.522268096645738, + "grad_norm": 0.0010161303216591477, + "learning_rate": 0.03476920294393337, + "loss": 0.084, + "num_input_tokens_seen": 40382400, + "step": 31155 + }, + { + "epoch": 1.5225123983094324, + "grad_norm": 0.0019672366324812174, + "learning_rate": 0.03473150072750755, + "loss": 0.0753, + "num_input_tokens_seen": 40388928, + "step": 31160 + }, + { + "epoch": 1.5227566999731268, + "grad_norm": 0.0013188289012759924, + "learning_rate": 0.03469381628693284, + "loss": 0.0726, + "num_input_tokens_seen": 40395808, + "step": 31165 + }, + { + "epoch": 1.5230010016368212, + "grad_norm": 0.0013061200734227896, + "learning_rate": 0.03465614962802072, + "loss": 0.075, + "num_input_tokens_seen": 40402016, + "step": 31170 + }, + { + "epoch": 1.5232453033005156, + "grad_norm": 0.0018879679264500737, + "learning_rate": 0.0346185007565798, + "loss": 0.0665, + "num_input_tokens_seen": 40408224, + "step": 31175 + }, + { + "epoch": 1.52348960496421, + "grad_norm": 0.001015312853269279, + "learning_rate": 0.03458086967841609, + "loss": 0.0901, + "num_input_tokens_seen": 40414368, + "step": 31180 + }, + { + "epoch": 1.523733906627904, + "grad_norm": 0.0012130577815696597, + "learning_rate": 0.03454325639933266, + "loss": 0.0705, + "num_input_tokens_seen": 40421152, + "step": 31185 + }, + { + "epoch": 1.5239782082915985, + "grad_norm": 0.001738836755976081, + "learning_rate": 0.03450566092513007, + "loss": 0.0835, + "num_input_tokens_seen": 40427072, + "step": 31190 + }, + { + "epoch": 1.5242225099552928, + "grad_norm": 0.0021388994064182043, + "learning_rate": 0.034468083261605914, + "loss": 0.0846, + "num_input_tokens_seen": 40433376, + "step": 31195 + }, + { + "epoch": 1.524466811618987, + "grad_norm": 0.0013015330769121647, + "learning_rate": 0.03443052341455522, + "loss": 0.083, + "num_input_tokens_seen": 40439712, + "step": 31200 + }, + { + "epoch": 1.524466811618987, + "eval_loss": 0.08627531677484512, + "eval_runtime": 402.311, + "eval_samples_per_second": 90.44, + "eval_steps_per_second": 22.612, + "num_input_tokens_seen": 40439712, + "step": 31200 + }, + { + "epoch": 1.5247111132826814, + "grad_norm": 0.0016315794782713056, + "learning_rate": 0.0343929813897701, + "loss": 0.0794, + "num_input_tokens_seen": 40445856, + "step": 31205 + }, + { + "epoch": 1.5249554149463758, + "grad_norm": 0.001072322716936469, + "learning_rate": 0.034355457193040125, + "loss": 0.0893, + "num_input_tokens_seen": 40452576, + "step": 31210 + }, + { + "epoch": 1.5251997166100701, + "grad_norm": 0.0013790144585072994, + "learning_rate": 0.03431795083015186, + "loss": 0.0766, + "num_input_tokens_seen": 40459296, + "step": 31215 + }, + { + "epoch": 1.5254440182737645, + "grad_norm": 0.001780350343324244, + "learning_rate": 0.03428046230688936, + "loss": 0.0861, + "num_input_tokens_seen": 40465536, + "step": 31220 + }, + { + "epoch": 1.525688319937459, + "grad_norm": 0.0009371057967655361, + "learning_rate": 0.034242991629033805, + "loss": 0.0784, + "num_input_tokens_seen": 40472096, + "step": 31225 + }, + { + "epoch": 1.525932621601153, + "grad_norm": 0.0007145963609218597, + "learning_rate": 0.03420553880236362, + "loss": 0.0677, + "num_input_tokens_seen": 40478528, + "step": 31230 + }, + { + "epoch": 1.5261769232648474, + "grad_norm": 0.001391104538924992, + "learning_rate": 0.03416810383265449, + "loss": 0.0893, + "num_input_tokens_seen": 40484512, + "step": 31235 + }, + { + "epoch": 1.5264212249285416, + "grad_norm": 0.0010480763157829642, + "learning_rate": 0.03413068672567944, + "loss": 0.0632, + "num_input_tokens_seen": 40491424, + "step": 31240 + }, + { + "epoch": 1.526665526592236, + "grad_norm": 0.0012331707403063774, + "learning_rate": 0.034093287487208565, + "loss": 0.0776, + "num_input_tokens_seen": 40497696, + "step": 31245 + }, + { + "epoch": 1.5269098282559304, + "grad_norm": 0.0017005964182317257, + "learning_rate": 0.03405590612300937, + "loss": 0.0994, + "num_input_tokens_seen": 40503872, + "step": 31250 + }, + { + "epoch": 1.5271541299196247, + "grad_norm": 0.0009671308216638863, + "learning_rate": 0.03401854263884646, + "loss": 0.08, + "num_input_tokens_seen": 40510176, + "step": 31255 + }, + { + "epoch": 1.5273984315833191, + "grad_norm": 0.0012122626649215817, + "learning_rate": 0.033981197040481824, + "loss": 0.0944, + "num_input_tokens_seen": 40516832, + "step": 31260 + }, + { + "epoch": 1.5276427332470135, + "grad_norm": 0.000955727300606668, + "learning_rate": 0.03394386933367459, + "loss": 0.063, + "num_input_tokens_seen": 40523104, + "step": 31265 + }, + { + "epoch": 1.5278870349107079, + "grad_norm": 0.0016708021285012364, + "learning_rate": 0.033906559524181104, + "loss": 0.1207, + "num_input_tokens_seen": 40529056, + "step": 31270 + }, + { + "epoch": 1.528131336574402, + "grad_norm": 0.0011528360191732645, + "learning_rate": 0.033869267617755085, + "loss": 0.0749, + "num_input_tokens_seen": 40535136, + "step": 31275 + }, + { + "epoch": 1.5283756382380964, + "grad_norm": 0.00123376096598804, + "learning_rate": 0.0338319936201474, + "loss": 0.0977, + "num_input_tokens_seen": 40542016, + "step": 31280 + }, + { + "epoch": 1.5286199399017906, + "grad_norm": 0.0015562777407467365, + "learning_rate": 0.033794737537106136, + "loss": 0.1171, + "num_input_tokens_seen": 40548448, + "step": 31285 + }, + { + "epoch": 1.528864241565485, + "grad_norm": 0.0010340482695028186, + "learning_rate": 0.03375749937437671, + "loss": 0.0583, + "num_input_tokens_seen": 40555008, + "step": 31290 + }, + { + "epoch": 1.5291085432291793, + "grad_norm": 0.0015529629308730364, + "learning_rate": 0.033720279137701634, + "loss": 0.0865, + "num_input_tokens_seen": 40561472, + "step": 31295 + }, + { + "epoch": 1.5293528448928737, + "grad_norm": 0.0015751280589029193, + "learning_rate": 0.03368307683282078, + "loss": 0.0672, + "num_input_tokens_seen": 40567936, + "step": 31300 + }, + { + "epoch": 1.529597146556568, + "grad_norm": 0.001714590354822576, + "learning_rate": 0.033645892465471235, + "loss": 0.081, + "num_input_tokens_seen": 40574912, + "step": 31305 + }, + { + "epoch": 1.5298414482202625, + "grad_norm": 0.0019970473367720842, + "learning_rate": 0.03360872604138724, + "loss": 0.1049, + "num_input_tokens_seen": 40581216, + "step": 31310 + }, + { + "epoch": 1.5300857498839568, + "grad_norm": 0.0013904586667194963, + "learning_rate": 0.03357157756630034, + "loss": 0.0816, + "num_input_tokens_seen": 40588032, + "step": 31315 + }, + { + "epoch": 1.530330051547651, + "grad_norm": 0.0018694453174248338, + "learning_rate": 0.033534447045939365, + "loss": 0.1225, + "num_input_tokens_seen": 40594432, + "step": 31320 + }, + { + "epoch": 1.5305743532113454, + "grad_norm": 0.0016862985212355852, + "learning_rate": 0.03349733448603026, + "loss": 0.0967, + "num_input_tokens_seen": 40600832, + "step": 31325 + }, + { + "epoch": 1.5308186548750395, + "grad_norm": 0.0012312072794884443, + "learning_rate": 0.03346023989229619, + "loss": 0.0732, + "num_input_tokens_seen": 40607488, + "step": 31330 + }, + { + "epoch": 1.531062956538734, + "grad_norm": 0.001723647117614746, + "learning_rate": 0.03342316327045769, + "loss": 0.0961, + "num_input_tokens_seen": 40614432, + "step": 31335 + }, + { + "epoch": 1.5313072582024283, + "grad_norm": 0.0014547505415976048, + "learning_rate": 0.033386104626232385, + "loss": 0.0623, + "num_input_tokens_seen": 40621504, + "step": 31340 + }, + { + "epoch": 1.5315515598661227, + "grad_norm": 0.0011854602489620447, + "learning_rate": 0.03334906396533525, + "loss": 0.0803, + "num_input_tokens_seen": 40627872, + "step": 31345 + }, + { + "epoch": 1.531795861529817, + "grad_norm": 0.0014828111743554473, + "learning_rate": 0.033312041293478326, + "loss": 0.0904, + "num_input_tokens_seen": 40634048, + "step": 31350 + }, + { + "epoch": 1.5320401631935114, + "grad_norm": 0.0011130647035315633, + "learning_rate": 0.03327503661637103, + "loss": 0.0787, + "num_input_tokens_seen": 40640640, + "step": 31355 + }, + { + "epoch": 1.5322844648572058, + "grad_norm": 0.0012118811719119549, + "learning_rate": 0.03323804993971998, + "loss": 0.0896, + "num_input_tokens_seen": 40646944, + "step": 31360 + }, + { + "epoch": 1.5325287665209, + "grad_norm": 0.001425315742380917, + "learning_rate": 0.033201081269228924, + "loss": 0.0851, + "num_input_tokens_seen": 40652992, + "step": 31365 + }, + { + "epoch": 1.5327730681845944, + "grad_norm": 0.0009520589956082404, + "learning_rate": 0.03316413061059895, + "loss": 0.0662, + "num_input_tokens_seen": 40659584, + "step": 31370 + }, + { + "epoch": 1.5330173698482885, + "grad_norm": 0.001066393218934536, + "learning_rate": 0.03312719796952827, + "loss": 0.0787, + "num_input_tokens_seen": 40666272, + "step": 31375 + }, + { + "epoch": 1.533261671511983, + "grad_norm": 0.0007689775666221976, + "learning_rate": 0.03309028335171236, + "loss": 0.073, + "num_input_tokens_seen": 40673088, + "step": 31380 + }, + { + "epoch": 1.5335059731756773, + "grad_norm": 0.0015451309736818075, + "learning_rate": 0.03305338676284398, + "loss": 0.0995, + "num_input_tokens_seen": 40680864, + "step": 31385 + }, + { + "epoch": 1.5337502748393717, + "grad_norm": 0.0008467303705401719, + "learning_rate": 0.03301650820861296, + "loss": 0.0524, + "num_input_tokens_seen": 40687072, + "step": 31390 + }, + { + "epoch": 1.533994576503066, + "grad_norm": 0.001834424096159637, + "learning_rate": 0.03297964769470652, + "loss": 0.1102, + "num_input_tokens_seen": 40693632, + "step": 31395 + }, + { + "epoch": 1.5342388781667604, + "grad_norm": 0.0014379300409927964, + "learning_rate": 0.032942805226808945, + "loss": 0.0546, + "num_input_tokens_seen": 40700736, + "step": 31400 + }, + { + "epoch": 1.5342388781667604, + "eval_loss": 0.0860108733177185, + "eval_runtime": 402.0752, + "eval_samples_per_second": 90.493, + "eval_steps_per_second": 22.625, + "num_input_tokens_seen": 40700736, + "step": 31400 + }, + { + "epoch": 1.5344831798304548, + "grad_norm": 0.0019913222640752792, + "learning_rate": 0.03290598081060187, + "loss": 0.071, + "num_input_tokens_seen": 40707424, + "step": 31405 + }, + { + "epoch": 1.534727481494149, + "grad_norm": 0.0014107811730355024, + "learning_rate": 0.03286917445176407, + "loss": 0.0891, + "num_input_tokens_seen": 40714400, + "step": 31410 + }, + { + "epoch": 1.5349717831578433, + "grad_norm": 0.0010952134616672993, + "learning_rate": 0.032832386155971456, + "loss": 0.0524, + "num_input_tokens_seen": 40720928, + "step": 31415 + }, + { + "epoch": 1.5352160848215375, + "grad_norm": 0.000923281186260283, + "learning_rate": 0.032795615928897334, + "loss": 0.0624, + "num_input_tokens_seen": 40727680, + "step": 31420 + }, + { + "epoch": 1.5354603864852319, + "grad_norm": 0.001117575797252357, + "learning_rate": 0.03275886377621215, + "loss": 0.077, + "num_input_tokens_seen": 40733888, + "step": 31425 + }, + { + "epoch": 1.5357046881489262, + "grad_norm": 0.0011854249751195312, + "learning_rate": 0.03272212970358348, + "loss": 0.0903, + "num_input_tokens_seen": 40740928, + "step": 31430 + }, + { + "epoch": 1.5359489898126206, + "grad_norm": 0.0016306600300595164, + "learning_rate": 0.032685413716676215, + "loss": 0.091, + "num_input_tokens_seen": 40747584, + "step": 31435 + }, + { + "epoch": 1.536193291476315, + "grad_norm": 0.0014444198459386826, + "learning_rate": 0.032648715821152474, + "loss": 0.0658, + "num_input_tokens_seen": 40753888, + "step": 31440 + }, + { + "epoch": 1.5364375931400094, + "grad_norm": 0.003391338512301445, + "learning_rate": 0.03261203602267143, + "loss": 0.089, + "num_input_tokens_seen": 40760192, + "step": 31445 + }, + { + "epoch": 1.5366818948037038, + "grad_norm": 0.0029305440839380026, + "learning_rate": 0.03257537432688966, + "loss": 0.0998, + "num_input_tokens_seen": 40766880, + "step": 31450 + }, + { + "epoch": 1.536926196467398, + "grad_norm": 0.001161741092801094, + "learning_rate": 0.03253873073946077, + "loss": 0.0777, + "num_input_tokens_seen": 40773056, + "step": 31455 + }, + { + "epoch": 1.5371704981310923, + "grad_norm": 0.0020205082837492228, + "learning_rate": 0.03250210526603572, + "loss": 0.092, + "num_input_tokens_seen": 40779136, + "step": 31460 + }, + { + "epoch": 1.5374147997947865, + "grad_norm": 0.001132558798417449, + "learning_rate": 0.03246549791226266, + "loss": 0.0824, + "num_input_tokens_seen": 40785792, + "step": 31465 + }, + { + "epoch": 1.5376591014584808, + "grad_norm": 0.0009813824435696006, + "learning_rate": 0.03242890868378679, + "loss": 0.0885, + "num_input_tokens_seen": 40792032, + "step": 31470 + }, + { + "epoch": 1.5379034031221752, + "grad_norm": 0.0014007706195116043, + "learning_rate": 0.03239233758625074, + "loss": 0.0882, + "num_input_tokens_seen": 40798688, + "step": 31475 + }, + { + "epoch": 1.5381477047858696, + "grad_norm": 0.0015446997713297606, + "learning_rate": 0.032355784625294204, + "loss": 0.0945, + "num_input_tokens_seen": 40805504, + "step": 31480 + }, + { + "epoch": 1.538392006449564, + "grad_norm": 0.001349360914900899, + "learning_rate": 0.03231924980655402, + "loss": 0.0829, + "num_input_tokens_seen": 40813312, + "step": 31485 + }, + { + "epoch": 1.5386363081132584, + "grad_norm": 0.0015588082605972886, + "learning_rate": 0.032282733135664446, + "loss": 0.0938, + "num_input_tokens_seen": 40819840, + "step": 31490 + }, + { + "epoch": 1.5388806097769527, + "grad_norm": 0.0016140613006427884, + "learning_rate": 0.03224623461825669, + "loss": 0.0752, + "num_input_tokens_seen": 40826432, + "step": 31495 + }, + { + "epoch": 1.539124911440647, + "grad_norm": 0.0012375491205602884, + "learning_rate": 0.03220975425995937, + "loss": 0.1092, + "num_input_tokens_seen": 40832992, + "step": 31500 + }, + { + "epoch": 1.5393692131043413, + "grad_norm": 0.0016187666915357113, + "learning_rate": 0.032173292066398206, + "loss": 0.0715, + "num_input_tokens_seen": 40839232, + "step": 31505 + }, + { + "epoch": 1.5396135147680354, + "grad_norm": 0.0012790015898644924, + "learning_rate": 0.03213684804319606, + "loss": 0.0759, + "num_input_tokens_seen": 40845696, + "step": 31510 + }, + { + "epoch": 1.5398578164317298, + "grad_norm": 0.0010548102436587214, + "learning_rate": 0.03210042219597312, + "loss": 0.0864, + "num_input_tokens_seen": 40852096, + "step": 31515 + }, + { + "epoch": 1.5401021180954242, + "grad_norm": 0.000873851589858532, + "learning_rate": 0.03206401453034675, + "loss": 0.075, + "num_input_tokens_seen": 40858592, + "step": 31520 + }, + { + "epoch": 1.5403464197591186, + "grad_norm": 0.0009736496722325683, + "learning_rate": 0.03202762505193136, + "loss": 0.069, + "num_input_tokens_seen": 40865696, + "step": 31525 + }, + { + "epoch": 1.540590721422813, + "grad_norm": 0.0016256208764389157, + "learning_rate": 0.031991253766338754, + "loss": 0.0825, + "num_input_tokens_seen": 40872288, + "step": 31530 + }, + { + "epoch": 1.5408350230865073, + "grad_norm": 0.000849608622957021, + "learning_rate": 0.03195490067917778, + "loss": 0.0661, + "num_input_tokens_seen": 40878560, + "step": 31535 + }, + { + "epoch": 1.5410793247502017, + "grad_norm": 0.0009934492409229279, + "learning_rate": 0.03191856579605461, + "loss": 0.0759, + "num_input_tokens_seen": 40884800, + "step": 31540 + }, + { + "epoch": 1.5413236264138959, + "grad_norm": 0.002510929247364402, + "learning_rate": 0.031882249122572454, + "loss": 0.0884, + "num_input_tokens_seen": 40891552, + "step": 31545 + }, + { + "epoch": 1.5415679280775902, + "grad_norm": 0.0018928381614387035, + "learning_rate": 0.03184595066433188, + "loss": 0.0859, + "num_input_tokens_seen": 40897952, + "step": 31550 + }, + { + "epoch": 1.5418122297412844, + "grad_norm": 0.0020937940571457148, + "learning_rate": 0.03180967042693049, + "loss": 0.0932, + "num_input_tokens_seen": 40905088, + "step": 31555 + }, + { + "epoch": 1.5420565314049788, + "grad_norm": 0.0014344300143420696, + "learning_rate": 0.03177340841596323, + "loss": 0.0923, + "num_input_tokens_seen": 40911584, + "step": 31560 + }, + { + "epoch": 1.5423008330686732, + "grad_norm": 0.0009064124315045774, + "learning_rate": 0.03173716463702209, + "loss": 0.0755, + "num_input_tokens_seen": 40917920, + "step": 31565 + }, + { + "epoch": 1.5425451347323675, + "grad_norm": 0.001400108216330409, + "learning_rate": 0.03170093909569638, + "loss": 0.0877, + "num_input_tokens_seen": 40924288, + "step": 31570 + }, + { + "epoch": 1.542789436396062, + "grad_norm": 0.0015809773467481136, + "learning_rate": 0.03166473179757246, + "loss": 0.09, + "num_input_tokens_seen": 40930464, + "step": 31575 + }, + { + "epoch": 1.5430337380597563, + "grad_norm": 0.0010238632094115019, + "learning_rate": 0.031628542748234005, + "loss": 0.0695, + "num_input_tokens_seen": 40936960, + "step": 31580 + }, + { + "epoch": 1.5432780397234507, + "grad_norm": 0.000699669704772532, + "learning_rate": 0.03159237195326184, + "loss": 0.0711, + "num_input_tokens_seen": 40943392, + "step": 31585 + }, + { + "epoch": 1.5435223413871448, + "grad_norm": 0.0011328537948429585, + "learning_rate": 0.031556219418233875, + "loss": 0.0783, + "num_input_tokens_seen": 40950176, + "step": 31590 + }, + { + "epoch": 1.5437666430508392, + "grad_norm": 0.0011483525158837438, + "learning_rate": 0.03152008514872533, + "loss": 0.0542, + "num_input_tokens_seen": 40956736, + "step": 31595 + }, + { + "epoch": 1.5440109447145334, + "grad_norm": 0.0017075254581868649, + "learning_rate": 0.03148396915030862, + "loss": 0.1009, + "num_input_tokens_seen": 40963072, + "step": 31600 + }, + { + "epoch": 1.5440109447145334, + "eval_loss": 0.0859449952840805, + "eval_runtime": 402.2386, + "eval_samples_per_second": 90.456, + "eval_steps_per_second": 22.616, + "num_input_tokens_seen": 40963072, + "step": 31600 + }, + { + "epoch": 1.5442552463782278, + "grad_norm": 0.0006992496200837195, + "learning_rate": 0.03144787142855318, + "loss": 0.0913, + "num_input_tokens_seen": 40969952, + "step": 31605 + }, + { + "epoch": 1.5444995480419221, + "grad_norm": 0.0014832502929493785, + "learning_rate": 0.031411791989025835, + "loss": 0.0807, + "num_input_tokens_seen": 40976704, + "step": 31610 + }, + { + "epoch": 1.5447438497056165, + "grad_norm": 0.001176374265924096, + "learning_rate": 0.031375730837290394, + "loss": 0.0982, + "num_input_tokens_seen": 40982528, + "step": 31615 + }, + { + "epoch": 1.544988151369311, + "grad_norm": 0.0006842592265456915, + "learning_rate": 0.031339687978908015, + "loss": 0.0662, + "num_input_tokens_seen": 40989568, + "step": 31620 + }, + { + "epoch": 1.5452324530330053, + "grad_norm": 0.0010248692706227303, + "learning_rate": 0.03130366341943694, + "loss": 0.071, + "num_input_tokens_seen": 40996192, + "step": 31625 + }, + { + "epoch": 1.5454767546966997, + "grad_norm": 0.001311065279878676, + "learning_rate": 0.031267657164432555, + "loss": 0.1133, + "num_input_tokens_seen": 41002784, + "step": 31630 + }, + { + "epoch": 1.5457210563603938, + "grad_norm": 0.0011017449432983994, + "learning_rate": 0.03123166921944752, + "loss": 0.0862, + "num_input_tokens_seen": 41009440, + "step": 31635 + }, + { + "epoch": 1.5459653580240882, + "grad_norm": 0.002200518501922488, + "learning_rate": 0.031195699590031666, + "loss": 0.1059, + "num_input_tokens_seen": 41015456, + "step": 31640 + }, + { + "epoch": 1.5462096596877823, + "grad_norm": 0.0008245795615948737, + "learning_rate": 0.031159748281731885, + "loss": 0.076, + "num_input_tokens_seen": 41021664, + "step": 31645 + }, + { + "epoch": 1.5464539613514767, + "grad_norm": 0.0014699295861646533, + "learning_rate": 0.031123815300092394, + "loss": 0.118, + "num_input_tokens_seen": 41027776, + "step": 31650 + }, + { + "epoch": 1.546698263015171, + "grad_norm": 0.0012113997945562005, + "learning_rate": 0.031087900650654424, + "loss": 0.0922, + "num_input_tokens_seen": 41034880, + "step": 31655 + }, + { + "epoch": 1.5469425646788655, + "grad_norm": 0.0017599508864805102, + "learning_rate": 0.031052004338956534, + "loss": 0.0903, + "num_input_tokens_seen": 41041568, + "step": 31660 + }, + { + "epoch": 1.5471868663425599, + "grad_norm": 0.001137303770519793, + "learning_rate": 0.031016126370534407, + "loss": 0.0812, + "num_input_tokens_seen": 41047904, + "step": 31665 + }, + { + "epoch": 1.5474311680062542, + "grad_norm": 0.0011287075467407703, + "learning_rate": 0.030980266750920804, + "loss": 0.0921, + "num_input_tokens_seen": 41054240, + "step": 31670 + }, + { + "epoch": 1.5476754696699484, + "grad_norm": 0.0008394119213335216, + "learning_rate": 0.030944425485645747, + "loss": 0.075, + "num_input_tokens_seen": 41060640, + "step": 31675 + }, + { + "epoch": 1.5479197713336428, + "grad_norm": 0.002142243552953005, + "learning_rate": 0.03090860258023647, + "loss": 0.0877, + "num_input_tokens_seen": 41067040, + "step": 31680 + }, + { + "epoch": 1.5481640729973372, + "grad_norm": 0.0011274119606241584, + "learning_rate": 0.030872798040217236, + "loss": 0.0701, + "num_input_tokens_seen": 41073472, + "step": 31685 + }, + { + "epoch": 1.5484083746610313, + "grad_norm": 0.0011601822916418314, + "learning_rate": 0.03083701187110964, + "loss": 0.0849, + "num_input_tokens_seen": 41080288, + "step": 31690 + }, + { + "epoch": 1.5486526763247257, + "grad_norm": 0.001167501206509769, + "learning_rate": 0.030801244078432294, + "loss": 0.1032, + "num_input_tokens_seen": 41086528, + "step": 31695 + }, + { + "epoch": 1.54889697798842, + "grad_norm": 0.0022017606534063816, + "learning_rate": 0.030765494667701024, + "loss": 0.1033, + "num_input_tokens_seen": 41092768, + "step": 31700 + }, + { + "epoch": 1.5491412796521145, + "grad_norm": 0.0014142879517748952, + "learning_rate": 0.030729763644428913, + "loss": 0.1183, + "num_input_tokens_seen": 41099360, + "step": 31705 + }, + { + "epoch": 1.5493855813158088, + "grad_norm": 0.001682609785348177, + "learning_rate": 0.030694051014126048, + "loss": 0.0962, + "num_input_tokens_seen": 41106592, + "step": 31710 + }, + { + "epoch": 1.5496298829795032, + "grad_norm": 0.0012364663416519761, + "learning_rate": 0.030658356782299792, + "loss": 0.079, + "num_input_tokens_seen": 41113376, + "step": 31715 + }, + { + "epoch": 1.5498741846431974, + "grad_norm": 0.0009877107804641128, + "learning_rate": 0.030622680954454726, + "loss": 0.0865, + "num_input_tokens_seen": 41119776, + "step": 31720 + }, + { + "epoch": 1.5501184863068918, + "grad_norm": 0.0009009384084492922, + "learning_rate": 0.030587023536092398, + "loss": 0.086, + "num_input_tokens_seen": 41126208, + "step": 31725 + }, + { + "epoch": 1.5503627879705861, + "grad_norm": 0.0012380011612549424, + "learning_rate": 0.03055138453271171, + "loss": 0.1236, + "num_input_tokens_seen": 41132512, + "step": 31730 + }, + { + "epoch": 1.5506070896342803, + "grad_norm": 0.0010081485379487276, + "learning_rate": 0.03051576394980858, + "loss": 0.0812, + "num_input_tokens_seen": 41139136, + "step": 31735 + }, + { + "epoch": 1.5508513912979747, + "grad_norm": 0.0007478652987629175, + "learning_rate": 0.030480161792876187, + "loss": 0.0532, + "num_input_tokens_seen": 41146016, + "step": 31740 + }, + { + "epoch": 1.551095692961669, + "grad_norm": 0.0014048555167391896, + "learning_rate": 0.030444578067404846, + "loss": 0.0895, + "num_input_tokens_seen": 41152224, + "step": 31745 + }, + { + "epoch": 1.5513399946253634, + "grad_norm": 0.001758594298735261, + "learning_rate": 0.030409012778881975, + "loss": 0.0994, + "num_input_tokens_seen": 41158976, + "step": 31750 + }, + { + "epoch": 1.5515842962890578, + "grad_norm": 0.0018606478115543723, + "learning_rate": 0.030373465932792235, + "loss": 0.1145, + "num_input_tokens_seen": 41165376, + "step": 31755 + }, + { + "epoch": 1.5518285979527522, + "grad_norm": 0.001325175166130066, + "learning_rate": 0.030337937534617342, + "loss": 0.1036, + "num_input_tokens_seen": 41172320, + "step": 31760 + }, + { + "epoch": 1.5520728996164463, + "grad_norm": 0.0014310163678601384, + "learning_rate": 0.030302427589836277, + "loss": 0.0965, + "num_input_tokens_seen": 41178752, + "step": 31765 + }, + { + "epoch": 1.5523172012801407, + "grad_norm": 0.0014626123011112213, + "learning_rate": 0.030266936103925095, + "loss": 0.0771, + "num_input_tokens_seen": 41185312, + "step": 31770 + }, + { + "epoch": 1.5525615029438349, + "grad_norm": 0.0013165645068511367, + "learning_rate": 0.030231463082356982, + "loss": 0.0968, + "num_input_tokens_seen": 41191840, + "step": 31775 + }, + { + "epoch": 1.5528058046075293, + "grad_norm": 0.0012475675903260708, + "learning_rate": 0.030196008530602367, + "loss": 0.0868, + "num_input_tokens_seen": 41198208, + "step": 31780 + }, + { + "epoch": 1.5530501062712236, + "grad_norm": 0.0016105725662782788, + "learning_rate": 0.030160572454128842, + "loss": 0.0703, + "num_input_tokens_seen": 41204832, + "step": 31785 + }, + { + "epoch": 1.553294407934918, + "grad_norm": 0.0008651247480884194, + "learning_rate": 0.03012515485840098, + "loss": 0.0731, + "num_input_tokens_seen": 41211584, + "step": 31790 + }, + { + "epoch": 1.5535387095986124, + "grad_norm": 0.001485613640397787, + "learning_rate": 0.030089755748880734, + "loss": 0.1115, + "num_input_tokens_seen": 41218080, + "step": 31795 + }, + { + "epoch": 1.5537830112623068, + "grad_norm": 0.0013103652745485306, + "learning_rate": 0.030054375131027003, + "loss": 0.0774, + "num_input_tokens_seen": 41224800, + "step": 31800 + }, + { + "epoch": 1.5537830112623068, + "eval_loss": 0.0856202244758606, + "eval_runtime": 402.041, + "eval_samples_per_second": 90.501, + "eval_steps_per_second": 22.627, + "num_input_tokens_seen": 41224800, + "step": 31800 + }, + { + "epoch": 1.5540273129260012, + "grad_norm": 0.0008422939572483301, + "learning_rate": 0.030019013010295942, + "loss": 0.0774, + "num_input_tokens_seen": 41231680, + "step": 31805 + }, + { + "epoch": 1.5542716145896953, + "grad_norm": 0.001130976714193821, + "learning_rate": 0.029983669392140897, + "loss": 0.0743, + "num_input_tokens_seen": 41238784, + "step": 31810 + }, + { + "epoch": 1.5545159162533897, + "grad_norm": 0.001298298710025847, + "learning_rate": 0.029948344282012217, + "loss": 0.094, + "num_input_tokens_seen": 41245088, + "step": 31815 + }, + { + "epoch": 1.5547602179170839, + "grad_norm": 0.0008517098613083363, + "learning_rate": 0.029913037685357507, + "loss": 0.0797, + "num_input_tokens_seen": 41251520, + "step": 31820 + }, + { + "epoch": 1.5550045195807782, + "grad_norm": 0.0008914362988434732, + "learning_rate": 0.029877749607621528, + "loss": 0.0672, + "num_input_tokens_seen": 41258016, + "step": 31825 + }, + { + "epoch": 1.5552488212444726, + "grad_norm": 0.0016569097060710192, + "learning_rate": 0.029842480054246077, + "loss": 0.0856, + "num_input_tokens_seen": 41264768, + "step": 31830 + }, + { + "epoch": 1.555493122908167, + "grad_norm": 0.0008859600638970733, + "learning_rate": 0.02980722903067022, + "loss": 0.0698, + "num_input_tokens_seen": 41271264, + "step": 31835 + }, + { + "epoch": 1.5557374245718614, + "grad_norm": 0.0004616097721736878, + "learning_rate": 0.029771996542330113, + "loss": 0.0605, + "num_input_tokens_seen": 41277568, + "step": 31840 + }, + { + "epoch": 1.5559817262355558, + "grad_norm": 0.001467327238060534, + "learning_rate": 0.029736782594658954, + "loss": 0.0843, + "num_input_tokens_seen": 41283680, + "step": 31845 + }, + { + "epoch": 1.5562260278992501, + "grad_norm": 0.0011392578016966581, + "learning_rate": 0.029701587193087284, + "loss": 0.0735, + "num_input_tokens_seen": 41290720, + "step": 31850 + }, + { + "epoch": 1.5564703295629443, + "grad_norm": 0.0005372367450036108, + "learning_rate": 0.0296664103430426, + "loss": 0.0617, + "num_input_tokens_seen": 41297664, + "step": 31855 + }, + { + "epoch": 1.5567146312266387, + "grad_norm": 0.0010983450338244438, + "learning_rate": 0.029631252049949652, + "loss": 0.0763, + "num_input_tokens_seen": 41304032, + "step": 31860 + }, + { + "epoch": 1.5569589328903328, + "grad_norm": 0.0016239688266068697, + "learning_rate": 0.02959611231923031, + "loss": 0.0958, + "num_input_tokens_seen": 41310368, + "step": 31865 + }, + { + "epoch": 1.5572032345540272, + "grad_norm": 0.0017201268346980214, + "learning_rate": 0.029560991156303507, + "loss": 0.0796, + "num_input_tokens_seen": 41316736, + "step": 31870 + }, + { + "epoch": 1.5574475362177216, + "grad_norm": 0.0018439724808558822, + "learning_rate": 0.02952588856658544, + "loss": 0.0794, + "num_input_tokens_seen": 41323552, + "step": 31875 + }, + { + "epoch": 1.557691837881416, + "grad_norm": 0.0015436536632478237, + "learning_rate": 0.029490804555489296, + "loss": 0.0734, + "num_input_tokens_seen": 41330560, + "step": 31880 + }, + { + "epoch": 1.5579361395451103, + "grad_norm": 0.001722603803500533, + "learning_rate": 0.029455739128425484, + "loss": 0.0895, + "num_input_tokens_seen": 41336832, + "step": 31885 + }, + { + "epoch": 1.5581804412088047, + "grad_norm": 0.0018876707181334496, + "learning_rate": 0.029420692290801607, + "loss": 0.0781, + "num_input_tokens_seen": 41343360, + "step": 31890 + }, + { + "epoch": 1.558424742872499, + "grad_norm": 0.0008632439421489835, + "learning_rate": 0.02938566404802223, + "loss": 0.0956, + "num_input_tokens_seen": 41349856, + "step": 31895 + }, + { + "epoch": 1.5586690445361933, + "grad_norm": 0.001313221757300198, + "learning_rate": 0.029350654405489195, + "loss": 0.075, + "num_input_tokens_seen": 41356736, + "step": 31900 + }, + { + "epoch": 1.5589133461998876, + "grad_norm": 0.0011836355552077293, + "learning_rate": 0.02931566336860145, + "loss": 0.0611, + "num_input_tokens_seen": 41363488, + "step": 31905 + }, + { + "epoch": 1.5591576478635818, + "grad_norm": 0.0014057550579309464, + "learning_rate": 0.02928069094275505, + "loss": 0.0899, + "num_input_tokens_seen": 41370112, + "step": 31910 + }, + { + "epoch": 1.5594019495272762, + "grad_norm": 0.0012933145044371486, + "learning_rate": 0.02924573713334314, + "loss": 0.0791, + "num_input_tokens_seen": 41376736, + "step": 31915 + }, + { + "epoch": 1.5596462511909706, + "grad_norm": 0.0024407017044723034, + "learning_rate": 0.02921080194575603, + "loss": 0.0934, + "num_input_tokens_seen": 41383040, + "step": 31920 + }, + { + "epoch": 1.559890552854665, + "grad_norm": 0.0009123545605689287, + "learning_rate": 0.029175885385381177, + "loss": 0.0857, + "num_input_tokens_seen": 41389376, + "step": 31925 + }, + { + "epoch": 1.5601348545183593, + "grad_norm": 0.0016189637826755643, + "learning_rate": 0.029140987457603223, + "loss": 0.0848, + "num_input_tokens_seen": 41396000, + "step": 31930 + }, + { + "epoch": 1.5603791561820537, + "grad_norm": 0.0009108100202865899, + "learning_rate": 0.029106108167803763, + "loss": 0.0787, + "num_input_tokens_seen": 41402432, + "step": 31935 + }, + { + "epoch": 1.560623457845748, + "grad_norm": 0.0011964512523263693, + "learning_rate": 0.029071247521361674, + "loss": 0.0976, + "num_input_tokens_seen": 41408928, + "step": 31940 + }, + { + "epoch": 1.5608677595094422, + "grad_norm": 0.0016180846141651273, + "learning_rate": 0.029036405523652945, + "loss": 0.0838, + "num_input_tokens_seen": 41415296, + "step": 31945 + }, + { + "epoch": 1.5611120611731366, + "grad_norm": 0.0009567724773660302, + "learning_rate": 0.029001582180050577, + "loss": 0.0807, + "num_input_tokens_seen": 41421920, + "step": 31950 + }, + { + "epoch": 1.5613563628368308, + "grad_norm": 0.0015378230018541217, + "learning_rate": 0.02896677749592482, + "loss": 0.1012, + "num_input_tokens_seen": 41428160, + "step": 31955 + }, + { + "epoch": 1.5616006645005251, + "grad_norm": 0.0019625192508101463, + "learning_rate": 0.028931991476642938, + "loss": 0.0785, + "num_input_tokens_seen": 41434880, + "step": 31960 + }, + { + "epoch": 1.5618449661642195, + "grad_norm": 0.0019219251116737723, + "learning_rate": 0.028897224127569412, + "loss": 0.0911, + "num_input_tokens_seen": 41441632, + "step": 31965 + }, + { + "epoch": 1.562089267827914, + "grad_norm": 0.0015400007832795382, + "learning_rate": 0.028862475454065832, + "loss": 0.0808, + "num_input_tokens_seen": 41448256, + "step": 31970 + }, + { + "epoch": 1.5623335694916083, + "grad_norm": 0.0013142499374225736, + "learning_rate": 0.028827745461490806, + "loss": 0.0882, + "num_input_tokens_seen": 41454528, + "step": 31975 + }, + { + "epoch": 1.5625778711553027, + "grad_norm": 0.001049310085363686, + "learning_rate": 0.028793034155200212, + "loss": 0.1063, + "num_input_tokens_seen": 41460960, + "step": 31980 + }, + { + "epoch": 1.562822172818997, + "grad_norm": 0.0011428137077018619, + "learning_rate": 0.028758341540546944, + "loss": 0.1006, + "num_input_tokens_seen": 41467072, + "step": 31985 + }, + { + "epoch": 1.5630664744826912, + "grad_norm": 0.0012339507229626179, + "learning_rate": 0.02872366762288098, + "loss": 0.0768, + "num_input_tokens_seen": 41473440, + "step": 31990 + }, + { + "epoch": 1.5633107761463856, + "grad_norm": 0.0009935081470757723, + "learning_rate": 0.028689012407549567, + "loss": 0.0732, + "num_input_tokens_seen": 41479520, + "step": 31995 + }, + { + "epoch": 1.5635550778100797, + "grad_norm": 0.0009321215329691768, + "learning_rate": 0.028654375899896892, + "loss": 0.0765, + "num_input_tokens_seen": 41485536, + "step": 32000 + }, + { + "epoch": 1.5635550778100797, + "eval_loss": 0.08558771759271622, + "eval_runtime": 403.1932, + "eval_samples_per_second": 90.242, + "eval_steps_per_second": 22.562, + "num_input_tokens_seen": 41485536, + "step": 32000 + }, + { + "epoch": 1.5637993794737741, + "grad_norm": 0.001122487592510879, + "learning_rate": 0.02861975810526437, + "loss": 0.0856, + "num_input_tokens_seen": 41492256, + "step": 32005 + }, + { + "epoch": 1.5640436811374685, + "grad_norm": 0.0015100190648809075, + "learning_rate": 0.02858515902899056, + "loss": 0.096, + "num_input_tokens_seen": 41498624, + "step": 32010 + }, + { + "epoch": 1.5642879828011629, + "grad_norm": 0.0010824560886248946, + "learning_rate": 0.028550578676410976, + "loss": 0.0685, + "num_input_tokens_seen": 41505888, + "step": 32015 + }, + { + "epoch": 1.5645322844648573, + "grad_norm": 0.000935180636588484, + "learning_rate": 0.02851601705285837, + "loss": 0.0642, + "num_input_tokens_seen": 41512128, + "step": 32020 + }, + { + "epoch": 1.5647765861285516, + "grad_norm": 0.0017366297543048859, + "learning_rate": 0.028481474163662666, + "loss": 0.0969, + "num_input_tokens_seen": 41518112, + "step": 32025 + }, + { + "epoch": 1.565020887792246, + "grad_norm": 0.0012541661271825433, + "learning_rate": 0.028446950014150683, + "loss": 0.0857, + "num_input_tokens_seen": 41524800, + "step": 32030 + }, + { + "epoch": 1.5652651894559402, + "grad_norm": 0.002178124152123928, + "learning_rate": 0.028412444609646596, + "loss": 0.0864, + "num_input_tokens_seen": 41531392, + "step": 32035 + }, + { + "epoch": 1.5655094911196346, + "grad_norm": 0.001922656549140811, + "learning_rate": 0.028377957955471465, + "loss": 0.0885, + "num_input_tokens_seen": 41537600, + "step": 32040 + }, + { + "epoch": 1.5657537927833287, + "grad_norm": 0.002788633806630969, + "learning_rate": 0.0283434900569436, + "loss": 0.0952, + "num_input_tokens_seen": 41543968, + "step": 32045 + }, + { + "epoch": 1.565998094447023, + "grad_norm": 0.0015519579173997045, + "learning_rate": 0.028309040919378456, + "loss": 0.0702, + "num_input_tokens_seen": 41550240, + "step": 32050 + }, + { + "epoch": 1.5662423961107175, + "grad_norm": 0.0013232880737632513, + "learning_rate": 0.02827461054808848, + "loss": 0.0945, + "num_input_tokens_seen": 41556448, + "step": 32055 + }, + { + "epoch": 1.5664866977744119, + "grad_norm": 0.0017695948481559753, + "learning_rate": 0.028240198948383186, + "loss": 0.1209, + "num_input_tokens_seen": 41562624, + "step": 32060 + }, + { + "epoch": 1.5667309994381062, + "grad_norm": 0.0012816559756174684, + "learning_rate": 0.028205806125569402, + "loss": 0.0998, + "num_input_tokens_seen": 41568512, + "step": 32065 + }, + { + "epoch": 1.5669753011018006, + "grad_norm": 0.001132130273617804, + "learning_rate": 0.028171432084950834, + "loss": 0.0741, + "num_input_tokens_seen": 41574784, + "step": 32070 + }, + { + "epoch": 1.567219602765495, + "grad_norm": 0.001182669191621244, + "learning_rate": 0.028137076831828478, + "loss": 0.0877, + "num_input_tokens_seen": 41580896, + "step": 32075 + }, + { + "epoch": 1.5674639044291891, + "grad_norm": 0.0007922311197035015, + "learning_rate": 0.028102740371500238, + "loss": 0.0921, + "num_input_tokens_seen": 41587232, + "step": 32080 + }, + { + "epoch": 1.5677082060928835, + "grad_norm": 0.0014122853754088283, + "learning_rate": 0.0280684227092613, + "loss": 0.0995, + "num_input_tokens_seen": 41593824, + "step": 32085 + }, + { + "epoch": 1.5679525077565777, + "grad_norm": 0.0016782439779490232, + "learning_rate": 0.02803412385040392, + "loss": 0.0934, + "num_input_tokens_seen": 41600128, + "step": 32090 + }, + { + "epoch": 1.568196809420272, + "grad_norm": 0.0013157492503523827, + "learning_rate": 0.027999843800217306, + "loss": 0.063, + "num_input_tokens_seen": 41607136, + "step": 32095 + }, + { + "epoch": 1.5684411110839664, + "grad_norm": 0.0011971151689067483, + "learning_rate": 0.027965582563987932, + "loss": 0.0895, + "num_input_tokens_seen": 41613408, + "step": 32100 + }, + { + "epoch": 1.5686854127476608, + "grad_norm": 0.0013745055766776204, + "learning_rate": 0.027931340146999346, + "loss": 0.0992, + "num_input_tokens_seen": 41620064, + "step": 32105 + }, + { + "epoch": 1.5689297144113552, + "grad_norm": 0.000981685472652316, + "learning_rate": 0.02789711655453208, + "loss": 0.0713, + "num_input_tokens_seen": 41626976, + "step": 32110 + }, + { + "epoch": 1.5691740160750496, + "grad_norm": 0.00121452403254807, + "learning_rate": 0.02786291179186392, + "loss": 0.0629, + "num_input_tokens_seen": 41633024, + "step": 32115 + }, + { + "epoch": 1.569418317738744, + "grad_norm": 0.0016504451632499695, + "learning_rate": 0.02782872586426961, + "loss": 0.0817, + "num_input_tokens_seen": 41639584, + "step": 32120 + }, + { + "epoch": 1.5696626194024381, + "grad_norm": 0.0013658548705279827, + "learning_rate": 0.027794558777021083, + "loss": 0.0688, + "num_input_tokens_seen": 41646880, + "step": 32125 + }, + { + "epoch": 1.5699069210661325, + "grad_norm": 0.0011544708395376801, + "learning_rate": 0.02776041053538734, + "loss": 0.0705, + "num_input_tokens_seen": 41653088, + "step": 32130 + }, + { + "epoch": 1.5701512227298267, + "grad_norm": 0.0015571173280477524, + "learning_rate": 0.027726281144634407, + "loss": 0.1011, + "num_input_tokens_seen": 41659520, + "step": 32135 + }, + { + "epoch": 1.570395524393521, + "grad_norm": 0.0016943515511229634, + "learning_rate": 0.02769217061002552, + "loss": 0.0665, + "num_input_tokens_seen": 41666144, + "step": 32140 + }, + { + "epoch": 1.5706398260572154, + "grad_norm": 0.0013099872739985585, + "learning_rate": 0.027658078936820967, + "loss": 0.0739, + "num_input_tokens_seen": 41671904, + "step": 32145 + }, + { + "epoch": 1.5708841277209098, + "grad_norm": 0.001203359803184867, + "learning_rate": 0.02762400613027805, + "loss": 0.0866, + "num_input_tokens_seen": 41678560, + "step": 32150 + }, + { + "epoch": 1.5711284293846042, + "grad_norm": 0.0017147830221801996, + "learning_rate": 0.027589952195651295, + "loss": 0.0858, + "num_input_tokens_seen": 41685280, + "step": 32155 + }, + { + "epoch": 1.5713727310482986, + "grad_norm": 0.0015930572990328074, + "learning_rate": 0.027555917138192186, + "loss": 0.0652, + "num_input_tokens_seen": 41691616, + "step": 32160 + }, + { + "epoch": 1.571617032711993, + "grad_norm": 0.0012526147766038775, + "learning_rate": 0.027521900963149375, + "loss": 0.0712, + "num_input_tokens_seen": 41697984, + "step": 32165 + }, + { + "epoch": 1.571861334375687, + "grad_norm": 0.0007742740563116968, + "learning_rate": 0.027487903675768633, + "loss": 0.068, + "num_input_tokens_seen": 41704352, + "step": 32170 + }, + { + "epoch": 1.5721056360393815, + "grad_norm": 0.0010318130953237414, + "learning_rate": 0.027453925281292677, + "loss": 0.0719, + "num_input_tokens_seen": 41711008, + "step": 32175 + }, + { + "epoch": 1.5723499377030756, + "grad_norm": 0.0016685385489836335, + "learning_rate": 0.027419965784961475, + "loss": 0.0772, + "num_input_tokens_seen": 41717760, + "step": 32180 + }, + { + "epoch": 1.57259423936677, + "grad_norm": 0.0010541274677962065, + "learning_rate": 0.027386025192012015, + "loss": 0.0775, + "num_input_tokens_seen": 41723904, + "step": 32185 + }, + { + "epoch": 1.5728385410304644, + "grad_norm": 0.0008494790527038276, + "learning_rate": 0.027352103507678277, + "loss": 0.1076, + "num_input_tokens_seen": 41730592, + "step": 32190 + }, + { + "epoch": 1.5730828426941588, + "grad_norm": 0.0013996184570714831, + "learning_rate": 0.027318200737191527, + "loss": 0.0769, + "num_input_tokens_seen": 41737184, + "step": 32195 + }, + { + "epoch": 1.5733271443578531, + "grad_norm": 0.0011767800897359848, + "learning_rate": 0.027284316885779935, + "loss": 0.1073, + "num_input_tokens_seen": 41743456, + "step": 32200 + }, + { + "epoch": 1.5733271443578531, + "eval_loss": 0.08534189313650131, + "eval_runtime": 402.6765, + "eval_samples_per_second": 90.358, + "eval_steps_per_second": 22.591, + "num_input_tokens_seen": 41743456, + "step": 32200 + }, + { + "epoch": 1.5735714460215475, + "grad_norm": 0.0007911674911156297, + "learning_rate": 0.027250451958668785, + "loss": 0.072, + "num_input_tokens_seen": 41750208, + "step": 32205 + }, + { + "epoch": 1.5738157476852417, + "grad_norm": 0.00131585996132344, + "learning_rate": 0.027216605961080536, + "loss": 0.0641, + "num_input_tokens_seen": 41757312, + "step": 32210 + }, + { + "epoch": 1.574060049348936, + "grad_norm": 0.0011394135653972626, + "learning_rate": 0.02718277889823461, + "loss": 0.0946, + "num_input_tokens_seen": 41763936, + "step": 32215 + }, + { + "epoch": 1.5743043510126304, + "grad_norm": 0.0013419546885415912, + "learning_rate": 0.027148970775347604, + "loss": 0.0699, + "num_input_tokens_seen": 41770592, + "step": 32220 + }, + { + "epoch": 1.5745486526763246, + "grad_norm": 0.0015604918589815497, + "learning_rate": 0.027115181597633174, + "loss": 0.1049, + "num_input_tokens_seen": 41776832, + "step": 32225 + }, + { + "epoch": 1.574792954340019, + "grad_norm": 0.0010329909855499864, + "learning_rate": 0.027081411370301976, + "loss": 0.0896, + "num_input_tokens_seen": 41782816, + "step": 32230 + }, + { + "epoch": 1.5750372560037134, + "grad_norm": 0.0013295281678438187, + "learning_rate": 0.027047660098561875, + "loss": 0.1039, + "num_input_tokens_seen": 41789248, + "step": 32235 + }, + { + "epoch": 1.5752815576674077, + "grad_norm": 0.0010297410190105438, + "learning_rate": 0.02701392778761766, + "loss": 0.0654, + "num_input_tokens_seen": 41795936, + "step": 32240 + }, + { + "epoch": 1.5755258593311021, + "grad_norm": 0.001699823304079473, + "learning_rate": 0.02698021444267133, + "loss": 0.1188, + "num_input_tokens_seen": 41802560, + "step": 32245 + }, + { + "epoch": 1.5757701609947965, + "grad_norm": 0.0010998089564964175, + "learning_rate": 0.026946520068921915, + "loss": 0.0654, + "num_input_tokens_seen": 41809248, + "step": 32250 + }, + { + "epoch": 1.5760144626584907, + "grad_norm": 0.001534226699732244, + "learning_rate": 0.02691284467156547, + "loss": 0.0984, + "num_input_tokens_seen": 41815936, + "step": 32255 + }, + { + "epoch": 1.576258764322185, + "grad_norm": 0.0009541672770865262, + "learning_rate": 0.026879188255795182, + "loss": 0.0844, + "num_input_tokens_seen": 41822624, + "step": 32260 + }, + { + "epoch": 1.5765030659858794, + "grad_norm": 0.0009810308692976832, + "learning_rate": 0.026845550826801328, + "loss": 0.0564, + "num_input_tokens_seen": 41829152, + "step": 32265 + }, + { + "epoch": 1.5767473676495736, + "grad_norm": 0.0029303664341568947, + "learning_rate": 0.02681193238977121, + "loss": 0.0951, + "num_input_tokens_seen": 41835424, + "step": 32270 + }, + { + "epoch": 1.576991669313268, + "grad_norm": 0.0009851530194282532, + "learning_rate": 0.026778332949889145, + "loss": 0.0709, + "num_input_tokens_seen": 41842272, + "step": 32275 + }, + { + "epoch": 1.5772359709769623, + "grad_norm": 0.0019462605705484748, + "learning_rate": 0.026744752512336673, + "loss": 0.0993, + "num_input_tokens_seen": 41848768, + "step": 32280 + }, + { + "epoch": 1.5774802726406567, + "grad_norm": 0.0010842586634680629, + "learning_rate": 0.02671119108229225, + "loss": 0.0959, + "num_input_tokens_seen": 41854912, + "step": 32285 + }, + { + "epoch": 1.577724574304351, + "grad_norm": 0.0009705163538455963, + "learning_rate": 0.026677648664931556, + "loss": 0.0815, + "num_input_tokens_seen": 41861344, + "step": 32290 + }, + { + "epoch": 1.5779688759680455, + "grad_norm": 0.0014006574638187885, + "learning_rate": 0.026644125265427154, + "loss": 0.0845, + "num_input_tokens_seen": 41867520, + "step": 32295 + }, + { + "epoch": 1.5782131776317396, + "grad_norm": 0.001182827283628285, + "learning_rate": 0.026610620888948822, + "loss": 0.0585, + "num_input_tokens_seen": 41874656, + "step": 32300 + }, + { + "epoch": 1.578457479295434, + "grad_norm": 0.0029884029645472765, + "learning_rate": 0.026577135540663408, + "loss": 0.0833, + "num_input_tokens_seen": 41881248, + "step": 32305 + }, + { + "epoch": 1.5787017809591284, + "grad_norm": 0.0012842625146731734, + "learning_rate": 0.026543669225734673, + "loss": 0.0677, + "num_input_tokens_seen": 41888608, + "step": 32310 + }, + { + "epoch": 1.5789460826228225, + "grad_norm": 0.0008483583224005997, + "learning_rate": 0.02651022194932363, + "loss": 0.1055, + "num_input_tokens_seen": 41895232, + "step": 32315 + }, + { + "epoch": 1.579190384286517, + "grad_norm": 0.001281282282434404, + "learning_rate": 0.026476793716588194, + "loss": 0.0844, + "num_input_tokens_seen": 41902368, + "step": 32320 + }, + { + "epoch": 1.5794346859502113, + "grad_norm": 0.0010972722666338086, + "learning_rate": 0.026443384532683467, + "loss": 0.0657, + "num_input_tokens_seen": 41909152, + "step": 32325 + }, + { + "epoch": 1.5796789876139057, + "grad_norm": 0.0012304509291425347, + "learning_rate": 0.026409994402761584, + "loss": 0.0682, + "num_input_tokens_seen": 41915904, + "step": 32330 + }, + { + "epoch": 1.5799232892776, + "grad_norm": 0.001906454679556191, + "learning_rate": 0.026376623331971653, + "loss": 0.094, + "num_input_tokens_seen": 41922240, + "step": 32335 + }, + { + "epoch": 1.5801675909412944, + "grad_norm": 0.0010781461605802178, + "learning_rate": 0.026343271325459997, + "loss": 0.069, + "num_input_tokens_seen": 41928512, + "step": 32340 + }, + { + "epoch": 1.5804118926049886, + "grad_norm": 0.0009410715429112315, + "learning_rate": 0.02630993838836987, + "loss": 0.0694, + "num_input_tokens_seen": 41935104, + "step": 32345 + }, + { + "epoch": 1.580656194268683, + "grad_norm": 0.001639701658859849, + "learning_rate": 0.026276624525841584, + "loss": 0.0995, + "num_input_tokens_seen": 41941856, + "step": 32350 + }, + { + "epoch": 1.5809004959323771, + "grad_norm": 0.0019958962220698595, + "learning_rate": 0.026243329743012637, + "loss": 0.0934, + "num_input_tokens_seen": 41948672, + "step": 32355 + }, + { + "epoch": 1.5811447975960715, + "grad_norm": 0.0007587409345433116, + "learning_rate": 0.026210054045017438, + "loss": 0.0715, + "num_input_tokens_seen": 41955232, + "step": 32360 + }, + { + "epoch": 1.581389099259766, + "grad_norm": 0.001278779236599803, + "learning_rate": 0.02617679743698755, + "loss": 0.0804, + "num_input_tokens_seen": 41961472, + "step": 32365 + }, + { + "epoch": 1.5816334009234603, + "grad_norm": 0.0009250518050976098, + "learning_rate": 0.02614355992405158, + "loss": 0.0984, + "num_input_tokens_seen": 41967424, + "step": 32370 + }, + { + "epoch": 1.5818777025871547, + "grad_norm": 0.0009360617841593921, + "learning_rate": 0.026110341511335115, + "loss": 0.0826, + "num_input_tokens_seen": 41974048, + "step": 32375 + }, + { + "epoch": 1.582122004250849, + "grad_norm": 0.0010192274348810315, + "learning_rate": 0.02607714220396093, + "loss": 0.0779, + "num_input_tokens_seen": 41980096, + "step": 32380 + }, + { + "epoch": 1.5823663059145434, + "grad_norm": 0.001103901886381209, + "learning_rate": 0.02604396200704869, + "loss": 0.0642, + "num_input_tokens_seen": 41986624, + "step": 32385 + }, + { + "epoch": 1.5826106075782376, + "grad_norm": 0.0012859405251219869, + "learning_rate": 0.02601080092571523, + "loss": 0.0755, + "num_input_tokens_seen": 41992832, + "step": 32390 + }, + { + "epoch": 1.582854909241932, + "grad_norm": 0.0010985780972987413, + "learning_rate": 0.025977658965074455, + "loss": 0.0715, + "num_input_tokens_seen": 41999328, + "step": 32395 + }, + { + "epoch": 1.583099210905626, + "grad_norm": 0.0010960969375446439, + "learning_rate": 0.02594453613023719, + "loss": 0.0824, + "num_input_tokens_seen": 42005696, + "step": 32400 + }, + { + "epoch": 1.583099210905626, + "eval_loss": 0.08598441630601883, + "eval_runtime": 402.4956, + "eval_samples_per_second": 90.398, + "eval_steps_per_second": 22.601, + "num_input_tokens_seen": 42005696, + "step": 32400 + }, + { + "epoch": 1.5833435125693205, + "grad_norm": 0.0015112695982679725, + "learning_rate": 0.025911432426311443, + "loss": 0.0759, + "num_input_tokens_seen": 42011808, + "step": 32405 + }, + { + "epoch": 1.5835878142330149, + "grad_norm": 0.0022349690552800894, + "learning_rate": 0.025878347858402234, + "loss": 0.0813, + "num_input_tokens_seen": 42018240, + "step": 32410 + }, + { + "epoch": 1.5838321158967092, + "grad_norm": 0.0016553864115849137, + "learning_rate": 0.025845282431611598, + "loss": 0.0891, + "num_input_tokens_seen": 42024800, + "step": 32415 + }, + { + "epoch": 1.5840764175604036, + "grad_norm": 0.0010850204853340983, + "learning_rate": 0.025812236151038608, + "loss": 0.0821, + "num_input_tokens_seen": 42031232, + "step": 32420 + }, + { + "epoch": 1.584320719224098, + "grad_norm": 0.0007904147496446967, + "learning_rate": 0.025779209021779468, + "loss": 0.0653, + "num_input_tokens_seen": 42037568, + "step": 32425 + }, + { + "epoch": 1.5845650208877924, + "grad_norm": 0.0008839526562951505, + "learning_rate": 0.025746201048927324, + "loss": 0.0704, + "num_input_tokens_seen": 42043872, + "step": 32430 + }, + { + "epoch": 1.5848093225514865, + "grad_norm": 0.0014206117484718561, + "learning_rate": 0.025713212237572485, + "loss": 0.0857, + "num_input_tokens_seen": 42050368, + "step": 32435 + }, + { + "epoch": 1.585053624215181, + "grad_norm": 0.0008773656445555389, + "learning_rate": 0.025680242592802164, + "loss": 0.0829, + "num_input_tokens_seen": 42056896, + "step": 32440 + }, + { + "epoch": 1.585297925878875, + "grad_norm": 0.0016514149028807878, + "learning_rate": 0.02564729211970073, + "loss": 0.0985, + "num_input_tokens_seen": 42063392, + "step": 32445 + }, + { + "epoch": 1.5855422275425695, + "grad_norm": 0.0016027624951675534, + "learning_rate": 0.025614360823349617, + "loss": 0.0839, + "num_input_tokens_seen": 42069824, + "step": 32450 + }, + { + "epoch": 1.5857865292062638, + "grad_norm": 0.0018984038615599275, + "learning_rate": 0.025581448708827146, + "loss": 0.1005, + "num_input_tokens_seen": 42076064, + "step": 32455 + }, + { + "epoch": 1.5860308308699582, + "grad_norm": 0.0013337020063772798, + "learning_rate": 0.025548555781208876, + "loss": 0.0746, + "num_input_tokens_seen": 42082784, + "step": 32460 + }, + { + "epoch": 1.5862751325336526, + "grad_norm": 0.000999462092295289, + "learning_rate": 0.02551568204556721, + "loss": 0.0767, + "num_input_tokens_seen": 42089248, + "step": 32465 + }, + { + "epoch": 1.586519434197347, + "grad_norm": 0.002073810435831547, + "learning_rate": 0.02548282750697173, + "loss": 0.0789, + "num_input_tokens_seen": 42095904, + "step": 32470 + }, + { + "epoch": 1.5867637358610414, + "grad_norm": 0.001604895107448101, + "learning_rate": 0.02544999217048909, + "loss": 0.0844, + "num_input_tokens_seen": 42102688, + "step": 32475 + }, + { + "epoch": 1.5870080375247355, + "grad_norm": 0.001007910119369626, + "learning_rate": 0.025417176041182793, + "loss": 0.0791, + "num_input_tokens_seen": 42108640, + "step": 32480 + }, + { + "epoch": 1.58725233918843, + "grad_norm": 0.001882205717265606, + "learning_rate": 0.025384379124113596, + "loss": 0.0854, + "num_input_tokens_seen": 42114720, + "step": 32485 + }, + { + "epoch": 1.587496640852124, + "grad_norm": 0.001891360036097467, + "learning_rate": 0.025351601424339124, + "loss": 0.1038, + "num_input_tokens_seen": 42121632, + "step": 32490 + }, + { + "epoch": 1.5877409425158184, + "grad_norm": 0.0010158636141568422, + "learning_rate": 0.025318842946914184, + "loss": 0.0778, + "num_input_tokens_seen": 42128320, + "step": 32495 + }, + { + "epoch": 1.5879852441795128, + "grad_norm": 0.0014928329037502408, + "learning_rate": 0.025286103696890494, + "loss": 0.0864, + "num_input_tokens_seen": 42134720, + "step": 32500 + }, + { + "epoch": 1.5882295458432072, + "grad_norm": 0.002266793278977275, + "learning_rate": 0.025253383679316836, + "loss": 0.1019, + "num_input_tokens_seen": 42141536, + "step": 32505 + }, + { + "epoch": 1.5884738475069016, + "grad_norm": 0.0018523283069953322, + "learning_rate": 0.025220682899239077, + "loss": 0.0874, + "num_input_tokens_seen": 42147584, + "step": 32510 + }, + { + "epoch": 1.588718149170596, + "grad_norm": 0.0012401459971442819, + "learning_rate": 0.02518800136170013, + "loss": 0.0691, + "num_input_tokens_seen": 42154080, + "step": 32515 + }, + { + "epoch": 1.5889624508342903, + "grad_norm": 0.0014921320835128427, + "learning_rate": 0.02515533907173981, + "loss": 0.0696, + "num_input_tokens_seen": 42161056, + "step": 32520 + }, + { + "epoch": 1.5892067524979845, + "grad_norm": 0.003225719090551138, + "learning_rate": 0.025122696034395115, + "loss": 0.1012, + "num_input_tokens_seen": 42167200, + "step": 32525 + }, + { + "epoch": 1.5894510541616789, + "grad_norm": 0.001268618507310748, + "learning_rate": 0.025090072254700023, + "loss": 0.0994, + "num_input_tokens_seen": 42174048, + "step": 32530 + }, + { + "epoch": 1.589695355825373, + "grad_norm": 0.001468090107664466, + "learning_rate": 0.025057467737685468, + "loss": 0.0655, + "num_input_tokens_seen": 42180320, + "step": 32535 + }, + { + "epoch": 1.5899396574890674, + "grad_norm": 0.0012067221105098724, + "learning_rate": 0.025024882488379557, + "loss": 0.0865, + "num_input_tokens_seen": 42187168, + "step": 32540 + }, + { + "epoch": 1.5901839591527618, + "grad_norm": 0.0009755706414580345, + "learning_rate": 0.02499231651180727, + "loss": 0.0731, + "num_input_tokens_seen": 42193952, + "step": 32545 + }, + { + "epoch": 1.5904282608164562, + "grad_norm": 0.0014490527100861073, + "learning_rate": 0.024959769812990713, + "loss": 0.0706, + "num_input_tokens_seen": 42200256, + "step": 32550 + }, + { + "epoch": 1.5906725624801505, + "grad_norm": 0.0008108124020509422, + "learning_rate": 0.024927242396949045, + "loss": 0.0547, + "num_input_tokens_seen": 42206528, + "step": 32555 + }, + { + "epoch": 1.590916864143845, + "grad_norm": 0.0011537218233570457, + "learning_rate": 0.02489473426869836, + "loss": 0.0798, + "num_input_tokens_seen": 42213280, + "step": 32560 + }, + { + "epoch": 1.5911611658075393, + "grad_norm": 0.001959435408934951, + "learning_rate": 0.024862245433251776, + "loss": 0.1044, + "num_input_tokens_seen": 42220128, + "step": 32565 + }, + { + "epoch": 1.5914054674712335, + "grad_norm": 0.002017243066802621, + "learning_rate": 0.024829775895619577, + "loss": 0.0695, + "num_input_tokens_seen": 42227104, + "step": 32570 + }, + { + "epoch": 1.5916497691349278, + "grad_norm": 0.0015124541241675615, + "learning_rate": 0.024797325660808882, + "loss": 0.0989, + "num_input_tokens_seen": 42233984, + "step": 32575 + }, + { + "epoch": 1.591894070798622, + "grad_norm": 0.0014151635114103556, + "learning_rate": 0.02476489473382401, + "loss": 0.0858, + "num_input_tokens_seen": 42240768, + "step": 32580 + }, + { + "epoch": 1.5921383724623164, + "grad_norm": 0.0014892774634063244, + "learning_rate": 0.024732483119666127, + "loss": 0.0726, + "num_input_tokens_seen": 42247712, + "step": 32585 + }, + { + "epoch": 1.5923826741260108, + "grad_norm": 0.0012443209998309612, + "learning_rate": 0.024700090823333548, + "loss": 0.0891, + "num_input_tokens_seen": 42254688, + "step": 32590 + }, + { + "epoch": 1.5926269757897051, + "grad_norm": 0.0008792077424004674, + "learning_rate": 0.02466771784982163, + "loss": 0.0891, + "num_input_tokens_seen": 42261120, + "step": 32595 + }, + { + "epoch": 1.5928712774533995, + "grad_norm": 0.0021307538263499737, + "learning_rate": 0.024635364204122594, + "loss": 0.1164, + "num_input_tokens_seen": 42267520, + "step": 32600 + }, + { + "epoch": 1.5928712774533995, + "eval_loss": 0.08523634076118469, + "eval_runtime": 402.6782, + "eval_samples_per_second": 90.358, + "eval_steps_per_second": 22.591, + "num_input_tokens_seen": 42267520, + "step": 32600 + }, + { + "epoch": 1.593115579117094, + "grad_norm": 0.0010123375104740262, + "learning_rate": 0.024603029891225852, + "loss": 0.07, + "num_input_tokens_seen": 42273952, + "step": 32605 + }, + { + "epoch": 1.5933598807807883, + "grad_norm": 0.0013731977669522166, + "learning_rate": 0.024570714916117748, + "loss": 0.1032, + "num_input_tokens_seen": 42280352, + "step": 32610 + }, + { + "epoch": 1.5936041824444824, + "grad_norm": 0.000929559872020036, + "learning_rate": 0.024538419283781625, + "loss": 0.0619, + "num_input_tokens_seen": 42286848, + "step": 32615 + }, + { + "epoch": 1.5938484841081768, + "grad_norm": 0.001125834882259369, + "learning_rate": 0.024506142999197938, + "loss": 0.0885, + "num_input_tokens_seen": 42293472, + "step": 32620 + }, + { + "epoch": 1.594092785771871, + "grad_norm": 0.0012323579285293818, + "learning_rate": 0.024473886067344002, + "loss": 0.08, + "num_input_tokens_seen": 42299552, + "step": 32625 + }, + { + "epoch": 1.5943370874355653, + "grad_norm": 0.001498354016803205, + "learning_rate": 0.02444164849319434, + "loss": 0.0891, + "num_input_tokens_seen": 42306432, + "step": 32630 + }, + { + "epoch": 1.5945813890992597, + "grad_norm": 0.0007634013891220093, + "learning_rate": 0.024409430281720306, + "loss": 0.0424, + "num_input_tokens_seen": 42313440, + "step": 32635 + }, + { + "epoch": 1.594825690762954, + "grad_norm": 0.0010932795703411102, + "learning_rate": 0.024377231437890428, + "loss": 0.0673, + "num_input_tokens_seen": 42319904, + "step": 32640 + }, + { + "epoch": 1.5950699924266485, + "grad_norm": 0.0014274305431172252, + "learning_rate": 0.024345051966670115, + "loss": 0.0991, + "num_input_tokens_seen": 42326272, + "step": 32645 + }, + { + "epoch": 1.5953142940903429, + "grad_norm": 0.0011182352900505066, + "learning_rate": 0.024312891873021884, + "loss": 0.096, + "num_input_tokens_seen": 42332384, + "step": 32650 + }, + { + "epoch": 1.5955585957540372, + "grad_norm": 0.0009002924198284745, + "learning_rate": 0.024280751161905183, + "loss": 0.0917, + "num_input_tokens_seen": 42338784, + "step": 32655 + }, + { + "epoch": 1.5958028974177314, + "grad_norm": 0.001220921752974391, + "learning_rate": 0.02424862983827658, + "loss": 0.0889, + "num_input_tokens_seen": 42345408, + "step": 32660 + }, + { + "epoch": 1.5960471990814258, + "grad_norm": 0.00158484757412225, + "learning_rate": 0.024216527907089495, + "loss": 0.0817, + "num_input_tokens_seen": 42351936, + "step": 32665 + }, + { + "epoch": 1.59629150074512, + "grad_norm": 0.0014037333894520998, + "learning_rate": 0.024184445373294505, + "loss": 0.0684, + "num_input_tokens_seen": 42358400, + "step": 32670 + }, + { + "epoch": 1.5965358024088143, + "grad_norm": 0.0015399408293887973, + "learning_rate": 0.02415238224183918, + "loss": 0.0947, + "num_input_tokens_seen": 42364224, + "step": 32675 + }, + { + "epoch": 1.5967801040725087, + "grad_norm": 0.00077469227835536, + "learning_rate": 0.024120338517667973, + "loss": 0.0643, + "num_input_tokens_seen": 42370816, + "step": 32680 + }, + { + "epoch": 1.597024405736203, + "grad_norm": 0.0013742136070504785, + "learning_rate": 0.02408831420572247, + "loss": 0.0853, + "num_input_tokens_seen": 42377440, + "step": 32685 + }, + { + "epoch": 1.5972687073998975, + "grad_norm": 0.001076096436008811, + "learning_rate": 0.024056309310941264, + "loss": 0.0716, + "num_input_tokens_seen": 42384128, + "step": 32690 + }, + { + "epoch": 1.5975130090635918, + "grad_norm": 0.0022730936761945486, + "learning_rate": 0.02402432383825982, + "loss": 0.0897, + "num_input_tokens_seen": 42390496, + "step": 32695 + }, + { + "epoch": 1.5977573107272862, + "grad_norm": 0.0014081299304962158, + "learning_rate": 0.023992357792610792, + "loss": 0.092, + "num_input_tokens_seen": 42397248, + "step": 32700 + }, + { + "epoch": 1.5980016123909804, + "grad_norm": 0.0010311590740457177, + "learning_rate": 0.0239604111789237, + "loss": 0.0659, + "num_input_tokens_seen": 42403520, + "step": 32705 + }, + { + "epoch": 1.5982459140546748, + "grad_norm": 0.0015301391249522567, + "learning_rate": 0.023928484002125095, + "loss": 0.0629, + "num_input_tokens_seen": 42410208, + "step": 32710 + }, + { + "epoch": 1.598490215718369, + "grad_norm": 0.0013139498187229037, + "learning_rate": 0.023896576267138595, + "loss": 0.0686, + "num_input_tokens_seen": 42416416, + "step": 32715 + }, + { + "epoch": 1.5987345173820633, + "grad_norm": 0.0011737875174731016, + "learning_rate": 0.02386468797888471, + "loss": 0.0902, + "num_input_tokens_seen": 42422528, + "step": 32720 + }, + { + "epoch": 1.5989788190457577, + "grad_norm": 0.0012521268799901009, + "learning_rate": 0.023832819142281057, + "loss": 0.0792, + "num_input_tokens_seen": 42428768, + "step": 32725 + }, + { + "epoch": 1.599223120709452, + "grad_norm": 0.0018440893618389964, + "learning_rate": 0.02380096976224225, + "loss": 0.086, + "num_input_tokens_seen": 42435552, + "step": 32730 + }, + { + "epoch": 1.5994674223731464, + "grad_norm": 0.0008585979812778533, + "learning_rate": 0.023769139843679777, + "loss": 0.0656, + "num_input_tokens_seen": 42442208, + "step": 32735 + }, + { + "epoch": 1.5997117240368408, + "grad_norm": 0.0013437046436592937, + "learning_rate": 0.023737329391502287, + "loss": 0.0757, + "num_input_tokens_seen": 42448960, + "step": 32740 + }, + { + "epoch": 1.599956025700535, + "grad_norm": 0.002041637198999524, + "learning_rate": 0.023705538410615293, + "loss": 0.0815, + "num_input_tokens_seen": 42455936, + "step": 32745 + }, + { + "epoch": 1.6002003273642293, + "grad_norm": 0.0010401641484349966, + "learning_rate": 0.023673766905921396, + "loss": 0.0588, + "num_input_tokens_seen": 42462528, + "step": 32750 + }, + { + "epoch": 1.6004446290279237, + "grad_norm": 0.0017249276861548424, + "learning_rate": 0.0236420148823202, + "loss": 0.0977, + "num_input_tokens_seen": 42469344, + "step": 32755 + }, + { + "epoch": 1.6006889306916179, + "grad_norm": 0.001553808804601431, + "learning_rate": 0.02361028234470816, + "loss": 0.1157, + "num_input_tokens_seen": 42475776, + "step": 32760 + }, + { + "epoch": 1.6009332323553123, + "grad_norm": 0.0026414792519062757, + "learning_rate": 0.023578569297978913, + "loss": 0.0969, + "num_input_tokens_seen": 42482112, + "step": 32765 + }, + { + "epoch": 1.6011775340190066, + "grad_norm": 0.0011370216961950064, + "learning_rate": 0.023546875747023025, + "loss": 0.0873, + "num_input_tokens_seen": 42488864, + "step": 32770 + }, + { + "epoch": 1.601421835682701, + "grad_norm": 0.001101392786949873, + "learning_rate": 0.02351520169672801, + "loss": 0.0666, + "num_input_tokens_seen": 42495232, + "step": 32775 + }, + { + "epoch": 1.6016661373463954, + "grad_norm": 0.0009562580962665379, + "learning_rate": 0.023483547151978357, + "loss": 0.0615, + "num_input_tokens_seen": 42501568, + "step": 32780 + }, + { + "epoch": 1.6019104390100898, + "grad_norm": 0.0014947887975722551, + "learning_rate": 0.023451912117655675, + "loss": 0.0949, + "num_input_tokens_seen": 42508032, + "step": 32785 + }, + { + "epoch": 1.602154740673784, + "grad_norm": 0.0011359038762748241, + "learning_rate": 0.023420296598638417, + "loss": 0.0682, + "num_input_tokens_seen": 42515360, + "step": 32790 + }, + { + "epoch": 1.6023990423374783, + "grad_norm": 0.0010764675680547953, + "learning_rate": 0.023388700599802165, + "loss": 0.0846, + "num_input_tokens_seen": 42522048, + "step": 32795 + }, + { + "epoch": 1.6026433440011727, + "grad_norm": 0.001500419806689024, + "learning_rate": 0.023357124126019334, + "loss": 0.0754, + "num_input_tokens_seen": 42528896, + "step": 32800 + }, + { + "epoch": 1.6026433440011727, + "eval_loss": 0.08678247779607773, + "eval_runtime": 402.9823, + "eval_samples_per_second": 90.289, + "eval_steps_per_second": 22.574, + "num_input_tokens_seen": 42528896, + "step": 32800 + }, + { + "epoch": 1.6028876456648669, + "grad_norm": 0.0008788650156930089, + "learning_rate": 0.02332556718215945, + "loss": 0.1149, + "num_input_tokens_seen": 42535008, + "step": 32805 + }, + { + "epoch": 1.6031319473285612, + "grad_norm": 0.0010586304124444723, + "learning_rate": 0.023294029773089035, + "loss": 0.0627, + "num_input_tokens_seen": 42541472, + "step": 32810 + }, + { + "epoch": 1.6033762489922556, + "grad_norm": 0.0007700303103774786, + "learning_rate": 0.023262511903671484, + "loss": 0.0514, + "num_input_tokens_seen": 42548096, + "step": 32815 + }, + { + "epoch": 1.60362055065595, + "grad_norm": 0.0019789261277765036, + "learning_rate": 0.023231013578767324, + "loss": 0.1275, + "num_input_tokens_seen": 42554688, + "step": 32820 + }, + { + "epoch": 1.6038648523196444, + "grad_norm": 0.0009556523291394114, + "learning_rate": 0.0231995348032339, + "loss": 0.0714, + "num_input_tokens_seen": 42561120, + "step": 32825 + }, + { + "epoch": 1.6041091539833388, + "grad_norm": 0.0012467378983274102, + "learning_rate": 0.023168075581925685, + "loss": 0.0737, + "num_input_tokens_seen": 42567744, + "step": 32830 + }, + { + "epoch": 1.604353455647033, + "grad_norm": 0.0018227691762149334, + "learning_rate": 0.023136635919694126, + "loss": 0.066, + "num_input_tokens_seen": 42574272, + "step": 32835 + }, + { + "epoch": 1.6045977573107273, + "grad_norm": 0.001391729572787881, + "learning_rate": 0.02310521582138753, + "loss": 0.0737, + "num_input_tokens_seen": 42581216, + "step": 32840 + }, + { + "epoch": 1.6048420589744217, + "grad_norm": 0.0009549366659484804, + "learning_rate": 0.023073815291851357, + "loss": 0.0778, + "num_input_tokens_seen": 42587200, + "step": 32845 + }, + { + "epoch": 1.6050863606381158, + "grad_norm": 0.0013245611917227507, + "learning_rate": 0.02304243433592788, + "loss": 0.0765, + "num_input_tokens_seen": 42593536, + "step": 32850 + }, + { + "epoch": 1.6053306623018102, + "grad_norm": 0.0015035680262371898, + "learning_rate": 0.023011072958456513, + "loss": 0.1084, + "num_input_tokens_seen": 42599904, + "step": 32855 + }, + { + "epoch": 1.6055749639655046, + "grad_norm": 0.0020564759615808725, + "learning_rate": 0.022979731164273536, + "loss": 0.0967, + "num_input_tokens_seen": 42606592, + "step": 32860 + }, + { + "epoch": 1.605819265629199, + "grad_norm": 0.0006792102358303964, + "learning_rate": 0.022948408958212218, + "loss": 0.0729, + "num_input_tokens_seen": 42612704, + "step": 32865 + }, + { + "epoch": 1.6060635672928933, + "grad_norm": 0.0012273831525817513, + "learning_rate": 0.022917106345102876, + "loss": 0.0761, + "num_input_tokens_seen": 42618816, + "step": 32870 + }, + { + "epoch": 1.6063078689565877, + "grad_norm": 0.0015480731381103396, + "learning_rate": 0.022885823329772785, + "loss": 0.0862, + "num_input_tokens_seen": 42625248, + "step": 32875 + }, + { + "epoch": 1.6065521706202819, + "grad_norm": 0.0013822570908814669, + "learning_rate": 0.02285455991704612, + "loss": 0.0862, + "num_input_tokens_seen": 42631488, + "step": 32880 + }, + { + "epoch": 1.6067964722839763, + "grad_norm": 0.0012002293951809406, + "learning_rate": 0.022823316111744117, + "loss": 0.0768, + "num_input_tokens_seen": 42637568, + "step": 32885 + }, + { + "epoch": 1.6070407739476704, + "grad_norm": 0.0012714724289253354, + "learning_rate": 0.022792091918685014, + "loss": 0.0717, + "num_input_tokens_seen": 42643936, + "step": 32890 + }, + { + "epoch": 1.6072850756113648, + "grad_norm": 0.00166858930606395, + "learning_rate": 0.022760887342683906, + "loss": 0.0777, + "num_input_tokens_seen": 42650432, + "step": 32895 + }, + { + "epoch": 1.6075293772750592, + "grad_norm": 0.0018929268699139357, + "learning_rate": 0.022729702388552975, + "loss": 0.0868, + "num_input_tokens_seen": 42657088, + "step": 32900 + }, + { + "epoch": 1.6077736789387536, + "grad_norm": 0.0010769619839265943, + "learning_rate": 0.022698537061101292, + "loss": 0.0757, + "num_input_tokens_seen": 42663840, + "step": 32905 + }, + { + "epoch": 1.608017980602448, + "grad_norm": 0.00112566736061126, + "learning_rate": 0.022667391365134962, + "loss": 0.0823, + "num_input_tokens_seen": 42670240, + "step": 32910 + }, + { + "epoch": 1.6082622822661423, + "grad_norm": 0.001174312550574541, + "learning_rate": 0.022636265305457065, + "loss": 0.0758, + "num_input_tokens_seen": 42676832, + "step": 32915 + }, + { + "epoch": 1.6085065839298367, + "grad_norm": 0.0009959266753867269, + "learning_rate": 0.02260515888686764, + "loss": 0.0864, + "num_input_tokens_seen": 42683232, + "step": 32920 + }, + { + "epoch": 1.6087508855935309, + "grad_norm": 0.001123466296121478, + "learning_rate": 0.022574072114163596, + "loss": 0.0656, + "num_input_tokens_seen": 42689920, + "step": 32925 + }, + { + "epoch": 1.6089951872572252, + "grad_norm": 0.001207459717988968, + "learning_rate": 0.022543004992139005, + "loss": 0.0803, + "num_input_tokens_seen": 42696416, + "step": 32930 + }, + { + "epoch": 1.6092394889209194, + "grad_norm": 0.001999286003410816, + "learning_rate": 0.022511957525584745, + "loss": 0.0909, + "num_input_tokens_seen": 42702656, + "step": 32935 + }, + { + "epoch": 1.6094837905846138, + "grad_norm": 0.0018538918811827898, + "learning_rate": 0.022480929719288778, + "loss": 0.1217, + "num_input_tokens_seen": 42708992, + "step": 32940 + }, + { + "epoch": 1.6097280922483082, + "grad_norm": 0.0013025462394580245, + "learning_rate": 0.02244992157803592, + "loss": 0.0746, + "num_input_tokens_seen": 42715168, + "step": 32945 + }, + { + "epoch": 1.6099723939120025, + "grad_norm": 0.0011379220522940159, + "learning_rate": 0.022418933106608047, + "loss": 0.0645, + "num_input_tokens_seen": 42721952, + "step": 32950 + }, + { + "epoch": 1.610216695575697, + "grad_norm": 0.0012237123446539044, + "learning_rate": 0.022387964309784018, + "loss": 0.0702, + "num_input_tokens_seen": 42728416, + "step": 32955 + }, + { + "epoch": 1.6104609972393913, + "grad_norm": 0.0013179947854951024, + "learning_rate": 0.022357015192339517, + "loss": 0.0779, + "num_input_tokens_seen": 42734848, + "step": 32960 + }, + { + "epoch": 1.6107052989030857, + "grad_norm": 0.0012192812282592058, + "learning_rate": 0.02232608575904734, + "loss": 0.0888, + "num_input_tokens_seen": 42741408, + "step": 32965 + }, + { + "epoch": 1.6109496005667798, + "grad_norm": 0.0014909112360328436, + "learning_rate": 0.022295176014677225, + "loss": 0.0774, + "num_input_tokens_seen": 42747616, + "step": 32970 + }, + { + "epoch": 1.6111939022304742, + "grad_norm": 0.0028550203423947096, + "learning_rate": 0.02226428596399577, + "loss": 0.11, + "num_input_tokens_seen": 42754144, + "step": 32975 + }, + { + "epoch": 1.6114382038941684, + "grad_norm": 0.001281941425986588, + "learning_rate": 0.02223341561176669, + "loss": 0.0764, + "num_input_tokens_seen": 42760448, + "step": 32980 + }, + { + "epoch": 1.6116825055578627, + "grad_norm": 0.0026859974022954702, + "learning_rate": 0.0222025649627505, + "loss": 0.0707, + "num_input_tokens_seen": 42766944, + "step": 32985 + }, + { + "epoch": 1.6119268072215571, + "grad_norm": 0.0009680322255007923, + "learning_rate": 0.022171734021704814, + "loss": 0.0832, + "num_input_tokens_seen": 42773376, + "step": 32990 + }, + { + "epoch": 1.6121711088852515, + "grad_norm": 0.0013759472640231252, + "learning_rate": 0.022140922793384116, + "loss": 0.0561, + "num_input_tokens_seen": 42780032, + "step": 32995 + }, + { + "epoch": 1.6124154105489459, + "grad_norm": 0.0018673575250431895, + "learning_rate": 0.022110131282539934, + "loss": 0.1035, + "num_input_tokens_seen": 42786240, + "step": 33000 + }, + { + "epoch": 1.6124154105489459, + "eval_loss": 0.08577519655227661, + "eval_runtime": 402.3989, + "eval_samples_per_second": 90.42, + "eval_steps_per_second": 22.607, + "num_input_tokens_seen": 42786240, + "step": 33000 + }, + { + "epoch": 1.6126597122126403, + "grad_norm": 0.001644924283027649, + "learning_rate": 0.022079359493920675, + "loss": 0.0801, + "num_input_tokens_seen": 42793216, + "step": 33005 + }, + { + "epoch": 1.6129040138763346, + "grad_norm": 0.0016620606184005737, + "learning_rate": 0.02204860743227169, + "loss": 0.0807, + "num_input_tokens_seen": 42799200, + "step": 33010 + }, + { + "epoch": 1.6131483155400288, + "grad_norm": 0.000985988648608327, + "learning_rate": 0.022017875102335365, + "loss": 0.0897, + "num_input_tokens_seen": 42805632, + "step": 33015 + }, + { + "epoch": 1.6133926172037232, + "grad_norm": 0.0015778786037117243, + "learning_rate": 0.02198716250885108, + "loss": 0.0876, + "num_input_tokens_seen": 42812160, + "step": 33020 + }, + { + "epoch": 1.6136369188674173, + "grad_norm": 0.0014078750973567367, + "learning_rate": 0.021956469656555, + "loss": 0.0673, + "num_input_tokens_seen": 42818848, + "step": 33025 + }, + { + "epoch": 1.6138812205311117, + "grad_norm": 0.001165946596302092, + "learning_rate": 0.0219257965501804, + "loss": 0.0954, + "num_input_tokens_seen": 42825024, + "step": 33030 + }, + { + "epoch": 1.614125522194806, + "grad_norm": 0.0017026803689077497, + "learning_rate": 0.021895143194457494, + "loss": 0.099, + "num_input_tokens_seen": 42831712, + "step": 33035 + }, + { + "epoch": 1.6143698238585005, + "grad_norm": 0.0014239627635106444, + "learning_rate": 0.021864509594113322, + "loss": 0.0695, + "num_input_tokens_seen": 42838176, + "step": 33040 + }, + { + "epoch": 1.6146141255221949, + "grad_norm": 0.0016438610618934035, + "learning_rate": 0.02183389575387207, + "loss": 0.0865, + "num_input_tokens_seen": 42845056, + "step": 33045 + }, + { + "epoch": 1.6148584271858892, + "grad_norm": 0.0012383876601234078, + "learning_rate": 0.021803301678454682, + "loss": 0.0854, + "num_input_tokens_seen": 42851456, + "step": 33050 + }, + { + "epoch": 1.6151027288495836, + "grad_norm": 0.0023073034826666117, + "learning_rate": 0.021772727372579213, + "loss": 0.0822, + "num_input_tokens_seen": 42857632, + "step": 33055 + }, + { + "epoch": 1.6153470305132778, + "grad_norm": 0.0014492328045889735, + "learning_rate": 0.02174217284096061, + "loss": 0.0936, + "num_input_tokens_seen": 42864320, + "step": 33060 + }, + { + "epoch": 1.6155913321769722, + "grad_norm": 0.00105315912514925, + "learning_rate": 0.0217116380883107, + "loss": 0.0972, + "num_input_tokens_seen": 42870720, + "step": 33065 + }, + { + "epoch": 1.6158356338406663, + "grad_norm": 0.0008671648683957756, + "learning_rate": 0.021681123119338425, + "loss": 0.0707, + "num_input_tokens_seen": 42876960, + "step": 33070 + }, + { + "epoch": 1.6160799355043607, + "grad_norm": 0.0011134914821013808, + "learning_rate": 0.02165062793874951, + "loss": 0.0954, + "num_input_tokens_seen": 42882976, + "step": 33075 + }, + { + "epoch": 1.616324237168055, + "grad_norm": 0.002059350488707423, + "learning_rate": 0.021620152551246666, + "loss": 0.0947, + "num_input_tokens_seen": 42889664, + "step": 33080 + }, + { + "epoch": 1.6165685388317494, + "grad_norm": 0.0008793145534582436, + "learning_rate": 0.02158969696152967, + "loss": 0.0803, + "num_input_tokens_seen": 42896192, + "step": 33085 + }, + { + "epoch": 1.6168128404954438, + "grad_norm": 0.0014247293584048748, + "learning_rate": 0.021559261174295057, + "loss": 0.0819, + "num_input_tokens_seen": 42902336, + "step": 33090 + }, + { + "epoch": 1.6170571421591382, + "grad_norm": 0.000771373335737735, + "learning_rate": 0.02152884519423646, + "loss": 0.0732, + "num_input_tokens_seen": 42909088, + "step": 33095 + }, + { + "epoch": 1.6173014438228326, + "grad_norm": 0.001376334112137556, + "learning_rate": 0.021498449026044447, + "loss": 0.0903, + "num_input_tokens_seen": 42916032, + "step": 33100 + }, + { + "epoch": 1.6175457454865267, + "grad_norm": 0.0009784310823306441, + "learning_rate": 0.021468072674406414, + "loss": 0.1087, + "num_input_tokens_seen": 42922272, + "step": 33105 + }, + { + "epoch": 1.6177900471502211, + "grad_norm": 0.001746227964758873, + "learning_rate": 0.021437716144006795, + "loss": 0.0942, + "num_input_tokens_seen": 42929056, + "step": 33110 + }, + { + "epoch": 1.6180343488139153, + "grad_norm": 0.0012062237365171313, + "learning_rate": 0.021407379439527002, + "loss": 0.0916, + "num_input_tokens_seen": 42935264, + "step": 33115 + }, + { + "epoch": 1.6182786504776097, + "grad_norm": 0.0009381069685332477, + "learning_rate": 0.021377062565645255, + "loss": 0.0624, + "num_input_tokens_seen": 42941920, + "step": 33120 + }, + { + "epoch": 1.618522952141304, + "grad_norm": 0.0022140436340123415, + "learning_rate": 0.02134676552703688, + "loss": 0.0581, + "num_input_tokens_seen": 42948352, + "step": 33125 + }, + { + "epoch": 1.6187672538049984, + "grad_norm": 0.00051040732068941, + "learning_rate": 0.02131648832837398, + "loss": 0.0501, + "num_input_tokens_seen": 42954336, + "step": 33130 + }, + { + "epoch": 1.6190115554686928, + "grad_norm": 0.0013705405872315168, + "learning_rate": 0.02128623097432574, + "loss": 0.0553, + "num_input_tokens_seen": 42960800, + "step": 33135 + }, + { + "epoch": 1.6192558571323872, + "grad_norm": 0.0015613092109560966, + "learning_rate": 0.021255993469558192, + "loss": 0.0856, + "num_input_tokens_seen": 42967392, + "step": 33140 + }, + { + "epoch": 1.6195001587960816, + "grad_norm": 0.0009908979991450906, + "learning_rate": 0.021225775818734364, + "loss": 0.0774, + "num_input_tokens_seen": 42973536, + "step": 33145 + }, + { + "epoch": 1.6197444604597757, + "grad_norm": 0.0012954124249517918, + "learning_rate": 0.021195578026514166, + "loss": 0.0765, + "num_input_tokens_seen": 42979808, + "step": 33150 + }, + { + "epoch": 1.61998876212347, + "grad_norm": 0.0008343150839209557, + "learning_rate": 0.02116540009755452, + "loss": 0.1184, + "num_input_tokens_seen": 42986368, + "step": 33155 + }, + { + "epoch": 1.6202330637871643, + "grad_norm": 0.0014934369828552008, + "learning_rate": 0.021135242036509173, + "loss": 0.0775, + "num_input_tokens_seen": 42993024, + "step": 33160 + }, + { + "epoch": 1.6204773654508586, + "grad_norm": 0.0018358161905780435, + "learning_rate": 0.021105103848028967, + "loss": 0.08, + "num_input_tokens_seen": 42999264, + "step": 33165 + }, + { + "epoch": 1.620721667114553, + "grad_norm": 0.0013662164565175772, + "learning_rate": 0.021074985536761504, + "loss": 0.0829, + "num_input_tokens_seen": 43005344, + "step": 33170 + }, + { + "epoch": 1.6209659687782474, + "grad_norm": 0.0016067902324721217, + "learning_rate": 0.021044887107351435, + "loss": 0.0878, + "num_input_tokens_seen": 43011680, + "step": 33175 + }, + { + "epoch": 1.6212102704419418, + "grad_norm": 0.0022051928099244833, + "learning_rate": 0.021014808564440362, + "loss": 0.0925, + "num_input_tokens_seen": 43017920, + "step": 33180 + }, + { + "epoch": 1.6214545721056361, + "grad_norm": 0.0016388407675549388, + "learning_rate": 0.02098474991266671, + "loss": 0.0776, + "num_input_tokens_seen": 43024512, + "step": 33185 + }, + { + "epoch": 1.6216988737693305, + "grad_norm": 0.0009871486108750105, + "learning_rate": 0.02095471115666592, + "loss": 0.0788, + "num_input_tokens_seen": 43030784, + "step": 33190 + }, + { + "epoch": 1.6219431754330247, + "grad_norm": 0.0016494861338287592, + "learning_rate": 0.020924692301070406, + "loss": 0.1039, + "num_input_tokens_seen": 43037184, + "step": 33195 + }, + { + "epoch": 1.622187477096719, + "grad_norm": 0.0009949073428288102, + "learning_rate": 0.020894693350509346, + "loss": 0.0824, + "num_input_tokens_seen": 43043616, + "step": 33200 + }, + { + "epoch": 1.622187477096719, + "eval_loss": 0.08585377037525177, + "eval_runtime": 402.3175, + "eval_samples_per_second": 90.439, + "eval_steps_per_second": 22.611, + "num_input_tokens_seen": 43043616, + "step": 33200 + }, + { + "epoch": 1.6224317787604132, + "grad_norm": 0.0006513723637908697, + "learning_rate": 0.020864714309609057, + "loss": 0.0809, + "num_input_tokens_seen": 43049824, + "step": 33205 + }, + { + "epoch": 1.6226760804241076, + "grad_norm": 0.001400183537043631, + "learning_rate": 0.020834755182992604, + "loss": 0.105, + "num_input_tokens_seen": 43056096, + "step": 33210 + }, + { + "epoch": 1.622920382087802, + "grad_norm": 0.0017605427419766784, + "learning_rate": 0.02080481597528011, + "loss": 0.1105, + "num_input_tokens_seen": 43062496, + "step": 33215 + }, + { + "epoch": 1.6231646837514964, + "grad_norm": 0.0009855934185907245, + "learning_rate": 0.020774896691088583, + "loss": 0.0693, + "num_input_tokens_seen": 43069216, + "step": 33220 + }, + { + "epoch": 1.6234089854151907, + "grad_norm": 0.0010030935518443584, + "learning_rate": 0.020744997335031882, + "loss": 0.0824, + "num_input_tokens_seen": 43075776, + "step": 33225 + }, + { + "epoch": 1.6236532870788851, + "grad_norm": 0.0013496197061613202, + "learning_rate": 0.02071511791172092, + "loss": 0.0777, + "num_input_tokens_seen": 43082496, + "step": 33230 + }, + { + "epoch": 1.6238975887425795, + "grad_norm": 0.001185797736980021, + "learning_rate": 0.02068525842576351, + "loss": 0.0813, + "num_input_tokens_seen": 43088608, + "step": 33235 + }, + { + "epoch": 1.6241418904062737, + "grad_norm": 0.0012045890325680375, + "learning_rate": 0.020655418881764264, + "loss": 0.0855, + "num_input_tokens_seen": 43095040, + "step": 33240 + }, + { + "epoch": 1.624386192069968, + "grad_norm": 0.0017967215972021222, + "learning_rate": 0.020625599284324923, + "loss": 0.0971, + "num_input_tokens_seen": 43101216, + "step": 33245 + }, + { + "epoch": 1.6246304937336622, + "grad_norm": 0.0004480066418182105, + "learning_rate": 0.02059579963804396, + "loss": 0.0454, + "num_input_tokens_seen": 43107840, + "step": 33250 + }, + { + "epoch": 1.6248747953973566, + "grad_norm": 0.0018007586477324367, + "learning_rate": 0.02056601994751688, + "loss": 0.0948, + "num_input_tokens_seen": 43114176, + "step": 33255 + }, + { + "epoch": 1.625119097061051, + "grad_norm": 0.0012802055571228266, + "learning_rate": 0.02053626021733614, + "loss": 0.0791, + "num_input_tokens_seen": 43120512, + "step": 33260 + }, + { + "epoch": 1.6253633987247453, + "grad_norm": 0.0014071835903450847, + "learning_rate": 0.02050652045209097, + "loss": 0.1126, + "num_input_tokens_seen": 43126688, + "step": 33265 + }, + { + "epoch": 1.6256077003884397, + "grad_norm": 0.0007258341065607965, + "learning_rate": 0.020476800656367672, + "loss": 0.0852, + "num_input_tokens_seen": 43132928, + "step": 33270 + }, + { + "epoch": 1.625852002052134, + "grad_norm": 0.0010987990535795689, + "learning_rate": 0.020447100834749425, + "loss": 0.0662, + "num_input_tokens_seen": 43138880, + "step": 33275 + }, + { + "epoch": 1.6260963037158283, + "grad_norm": 0.001995709026232362, + "learning_rate": 0.02041742099181627, + "loss": 0.0792, + "num_input_tokens_seen": 43145568, + "step": 33280 + }, + { + "epoch": 1.6263406053795226, + "grad_norm": 0.0012915136758238077, + "learning_rate": 0.02038776113214526, + "loss": 0.1112, + "num_input_tokens_seen": 43152160, + "step": 33285 + }, + { + "epoch": 1.626584907043217, + "grad_norm": 0.0021295377518981695, + "learning_rate": 0.0203581212603103, + "loss": 0.0767, + "num_input_tokens_seen": 43158880, + "step": 33290 + }, + { + "epoch": 1.6268292087069112, + "grad_norm": 0.001211784197948873, + "learning_rate": 0.02032850138088219, + "loss": 0.0662, + "num_input_tokens_seen": 43165440, + "step": 33295 + }, + { + "epoch": 1.6270735103706055, + "grad_norm": 0.0008618126739747822, + "learning_rate": 0.020298901498428754, + "loss": 0.0783, + "num_input_tokens_seen": 43171872, + "step": 33300 + }, + { + "epoch": 1.6273178120343, + "grad_norm": 0.0015580639010295272, + "learning_rate": 0.020269321617514595, + "loss": 0.1052, + "num_input_tokens_seen": 43178304, + "step": 33305 + }, + { + "epoch": 1.6275621136979943, + "grad_norm": 0.0021859128028154373, + "learning_rate": 0.020239761742701343, + "loss": 0.0877, + "num_input_tokens_seen": 43184320, + "step": 33310 + }, + { + "epoch": 1.6278064153616887, + "grad_norm": 0.0011176185216754675, + "learning_rate": 0.02021022187854754, + "loss": 0.0691, + "num_input_tokens_seen": 43190656, + "step": 33315 + }, + { + "epoch": 1.628050717025383, + "grad_norm": 0.0010279647540301085, + "learning_rate": 0.020180702029608522, + "loss": 0.0604, + "num_input_tokens_seen": 43197440, + "step": 33320 + }, + { + "epoch": 1.6282950186890772, + "grad_norm": 0.0008801005315035582, + "learning_rate": 0.020151202200436695, + "loss": 0.0594, + "num_input_tokens_seen": 43203712, + "step": 33325 + }, + { + "epoch": 1.6285393203527716, + "grad_norm": 0.0024007679894566536, + "learning_rate": 0.020121722395581226, + "loss": 0.0824, + "num_input_tokens_seen": 43210656, + "step": 33330 + }, + { + "epoch": 1.628783622016466, + "grad_norm": 0.0012218777555972338, + "learning_rate": 0.020092262619588342, + "loss": 0.0694, + "num_input_tokens_seen": 43217088, + "step": 33335 + }, + { + "epoch": 1.6290279236801601, + "grad_norm": 0.0007413559360429645, + "learning_rate": 0.02006282287700109, + "loss": 0.0703, + "num_input_tokens_seen": 43223712, + "step": 33340 + }, + { + "epoch": 1.6292722253438545, + "grad_norm": 0.001125496462918818, + "learning_rate": 0.020033403172359427, + "loss": 0.0864, + "num_input_tokens_seen": 43230432, + "step": 33345 + }, + { + "epoch": 1.629516527007549, + "grad_norm": 0.0010525438701733947, + "learning_rate": 0.020004003510200284, + "loss": 0.101, + "num_input_tokens_seen": 43236992, + "step": 33350 + }, + { + "epoch": 1.6297608286712433, + "grad_norm": 0.001461912994273007, + "learning_rate": 0.019974623895057407, + "loss": 0.0653, + "num_input_tokens_seen": 43243264, + "step": 33355 + }, + { + "epoch": 1.6300051303349377, + "grad_norm": 0.002836428117007017, + "learning_rate": 0.019945264331461553, + "loss": 0.1066, + "num_input_tokens_seen": 43249536, + "step": 33360 + }, + { + "epoch": 1.630249431998632, + "grad_norm": 0.0011474934872239828, + "learning_rate": 0.019915924823940317, + "loss": 0.0928, + "num_input_tokens_seen": 43256288, + "step": 33365 + }, + { + "epoch": 1.6304937336623262, + "grad_norm": 0.0012928448850288987, + "learning_rate": 0.01988660537701816, + "loss": 0.087, + "num_input_tokens_seen": 43263104, + "step": 33370 + }, + { + "epoch": 1.6307380353260206, + "grad_norm": 0.0013961032964289188, + "learning_rate": 0.01985730599521659, + "loss": 0.0521, + "num_input_tokens_seen": 43269408, + "step": 33375 + }, + { + "epoch": 1.630982336989715, + "grad_norm": 0.0015643546357750893, + "learning_rate": 0.019828026683053918, + "loss": 0.0916, + "num_input_tokens_seen": 43275936, + "step": 33380 + }, + { + "epoch": 1.6312266386534091, + "grad_norm": 0.0011630739318206906, + "learning_rate": 0.01979876744504535, + "loss": 0.081, + "num_input_tokens_seen": 43282208, + "step": 33385 + }, + { + "epoch": 1.6314709403171035, + "grad_norm": 0.0016958605265244842, + "learning_rate": 0.019769528285703046, + "loss": 0.0973, + "num_input_tokens_seen": 43288288, + "step": 33390 + }, + { + "epoch": 1.6317152419807979, + "grad_norm": 0.0017396204639226198, + "learning_rate": 0.019740309209536098, + "loss": 0.0748, + "num_input_tokens_seen": 43294400, + "step": 33395 + }, + { + "epoch": 1.6319595436444922, + "grad_norm": 0.0013015029253438115, + "learning_rate": 0.019711110221050387, + "loss": 0.085, + "num_input_tokens_seen": 43300896, + "step": 33400 + }, + { + "epoch": 1.6319595436444922, + "eval_loss": 0.08452259004116058, + "eval_runtime": 402.5956, + "eval_samples_per_second": 90.376, + "eval_steps_per_second": 22.596, + "num_input_tokens_seen": 43300896, + "step": 33400 + }, + { + "epoch": 1.6322038453081866, + "grad_norm": 0.0012868352932855487, + "learning_rate": 0.019681931324748825, + "loss": 0.079, + "num_input_tokens_seen": 43307520, + "step": 33405 + }, + { + "epoch": 1.632448146971881, + "grad_norm": 0.0011037400690838695, + "learning_rate": 0.019652772525131094, + "loss": 0.092, + "num_input_tokens_seen": 43313600, + "step": 33410 + }, + { + "epoch": 1.6326924486355752, + "grad_norm": 0.0011844569817185402, + "learning_rate": 0.019623633826693885, + "loss": 0.0741, + "num_input_tokens_seen": 43320160, + "step": 33415 + }, + { + "epoch": 1.6329367502992695, + "grad_norm": 0.000985751743428409, + "learning_rate": 0.019594515233930788, + "loss": 0.0622, + "num_input_tokens_seen": 43326560, + "step": 33420 + }, + { + "epoch": 1.6331810519629637, + "grad_norm": 0.0007357394206337631, + "learning_rate": 0.019565416751332186, + "loss": 0.0644, + "num_input_tokens_seen": 43332672, + "step": 33425 + }, + { + "epoch": 1.633425353626658, + "grad_norm": 0.001185859553515911, + "learning_rate": 0.019536338383385497, + "loss": 0.0717, + "num_input_tokens_seen": 43338880, + "step": 33430 + }, + { + "epoch": 1.6336696552903525, + "grad_norm": 0.0010740869911387563, + "learning_rate": 0.019507280134574933, + "loss": 0.0731, + "num_input_tokens_seen": 43345440, + "step": 33435 + }, + { + "epoch": 1.6339139569540468, + "grad_norm": 0.0013164934935048223, + "learning_rate": 0.019478242009381624, + "loss": 0.0775, + "num_input_tokens_seen": 43351936, + "step": 33440 + }, + { + "epoch": 1.6341582586177412, + "grad_norm": 0.0012434071395546198, + "learning_rate": 0.01944922401228367, + "loss": 0.0934, + "num_input_tokens_seen": 43358176, + "step": 33445 + }, + { + "epoch": 1.6344025602814356, + "grad_norm": 0.0013684541918337345, + "learning_rate": 0.01942022614775593, + "loss": 0.0932, + "num_input_tokens_seen": 43364544, + "step": 33450 + }, + { + "epoch": 1.63464686194513, + "grad_norm": 0.0014281062176451087, + "learning_rate": 0.01939124842027029, + "loss": 0.0982, + "num_input_tokens_seen": 43370912, + "step": 33455 + }, + { + "epoch": 1.6348911636088241, + "grad_norm": 0.0011501666158437729, + "learning_rate": 0.01936229083429551, + "loss": 0.0694, + "num_input_tokens_seen": 43377024, + "step": 33460 + }, + { + "epoch": 1.6351354652725185, + "grad_norm": 0.0014865434495732188, + "learning_rate": 0.019333353394297148, + "loss": 0.086, + "num_input_tokens_seen": 43383968, + "step": 33465 + }, + { + "epoch": 1.6353797669362127, + "grad_norm": 0.0011833782773464918, + "learning_rate": 0.019304436104737754, + "loss": 0.0977, + "num_input_tokens_seen": 43390528, + "step": 33470 + }, + { + "epoch": 1.635624068599907, + "grad_norm": 0.0010611076140776277, + "learning_rate": 0.019275538970076778, + "loss": 0.0625, + "num_input_tokens_seen": 43397184, + "step": 33475 + }, + { + "epoch": 1.6358683702636014, + "grad_norm": 0.0006797212990932167, + "learning_rate": 0.019246661994770434, + "loss": 0.0693, + "num_input_tokens_seen": 43403456, + "step": 33480 + }, + { + "epoch": 1.6361126719272958, + "grad_norm": 0.0009869454661384225, + "learning_rate": 0.019217805183271985, + "loss": 0.0707, + "num_input_tokens_seen": 43409760, + "step": 33485 + }, + { + "epoch": 1.6363569735909902, + "grad_norm": 0.001516828779131174, + "learning_rate": 0.019188968540031465, + "loss": 0.0699, + "num_input_tokens_seen": 43416800, + "step": 33490 + }, + { + "epoch": 1.6366012752546846, + "grad_norm": 0.001487785135395825, + "learning_rate": 0.019160152069495867, + "loss": 0.093, + "num_input_tokens_seen": 43423648, + "step": 33495 + }, + { + "epoch": 1.636845576918379, + "grad_norm": 0.001400655135512352, + "learning_rate": 0.019131355776109103, + "loss": 0.0724, + "num_input_tokens_seen": 43430240, + "step": 33500 + }, + { + "epoch": 1.637089878582073, + "grad_norm": 0.0013877853052690625, + "learning_rate": 0.019102579664311857, + "loss": 0.0766, + "num_input_tokens_seen": 43436832, + "step": 33505 + }, + { + "epoch": 1.6373341802457675, + "grad_norm": 0.0014536266680806875, + "learning_rate": 0.019073823738541763, + "loss": 0.0863, + "num_input_tokens_seen": 43442912, + "step": 33510 + }, + { + "epoch": 1.6375784819094616, + "grad_norm": 0.0011234376579523087, + "learning_rate": 0.0190450880032334, + "loss": 0.0872, + "num_input_tokens_seen": 43449408, + "step": 33515 + }, + { + "epoch": 1.637822783573156, + "grad_norm": 0.0008436717325821519, + "learning_rate": 0.019016372462818114, + "loss": 0.073, + "num_input_tokens_seen": 43456096, + "step": 33520 + }, + { + "epoch": 1.6380670852368504, + "grad_norm": 0.0013449689140543342, + "learning_rate": 0.018987677121724278, + "loss": 0.0983, + "num_input_tokens_seen": 43462528, + "step": 33525 + }, + { + "epoch": 1.6383113869005448, + "grad_norm": 0.001485709217377007, + "learning_rate": 0.018959001984377, + "loss": 0.1042, + "num_input_tokens_seen": 43469280, + "step": 33530 + }, + { + "epoch": 1.6385556885642392, + "grad_norm": 0.0011799648636952043, + "learning_rate": 0.018930347055198377, + "loss": 0.0892, + "num_input_tokens_seen": 43475616, + "step": 33535 + }, + { + "epoch": 1.6387999902279335, + "grad_norm": 0.0008549883496016264, + "learning_rate": 0.01890171233860739, + "loss": 0.0627, + "num_input_tokens_seen": 43482048, + "step": 33540 + }, + { + "epoch": 1.639044291891628, + "grad_norm": 0.0011961604468524456, + "learning_rate": 0.018873097839019807, + "loss": 0.0663, + "num_input_tokens_seen": 43488416, + "step": 33545 + }, + { + "epoch": 1.639288593555322, + "grad_norm": 0.0008709739777259529, + "learning_rate": 0.0188445035608484, + "loss": 0.0749, + "num_input_tokens_seen": 43495136, + "step": 33550 + }, + { + "epoch": 1.6395328952190165, + "grad_norm": 0.0019069879781454802, + "learning_rate": 0.018815929508502777, + "loss": 0.1077, + "num_input_tokens_seen": 43501472, + "step": 33555 + }, + { + "epoch": 1.6397771968827106, + "grad_norm": 0.0010383616900071502, + "learning_rate": 0.01878737568638934, + "loss": 0.0893, + "num_input_tokens_seen": 43508032, + "step": 33560 + }, + { + "epoch": 1.640021498546405, + "grad_norm": 0.0020060702227056026, + "learning_rate": 0.01875884209891152, + "loss": 0.0644, + "num_input_tokens_seen": 43514304, + "step": 33565 + }, + { + "epoch": 1.6402658002100994, + "grad_norm": 0.001086603500880301, + "learning_rate": 0.018730328750469514, + "loss": 0.0877, + "num_input_tokens_seen": 43520960, + "step": 33570 + }, + { + "epoch": 1.6405101018737938, + "grad_norm": 0.0017491549951955676, + "learning_rate": 0.018701835645460473, + "loss": 0.0757, + "num_input_tokens_seen": 43527040, + "step": 33575 + }, + { + "epoch": 1.6407544035374881, + "grad_norm": 0.0016593564068898559, + "learning_rate": 0.01867336278827838, + "loss": 0.0653, + "num_input_tokens_seen": 43533536, + "step": 33580 + }, + { + "epoch": 1.6409987052011825, + "grad_norm": 0.0008592510712333024, + "learning_rate": 0.018644910183314056, + "loss": 0.0607, + "num_input_tokens_seen": 43540288, + "step": 33585 + }, + { + "epoch": 1.641243006864877, + "grad_norm": 0.001065752119757235, + "learning_rate": 0.01861647783495531, + "loss": 0.0735, + "num_input_tokens_seen": 43546720, + "step": 33590 + }, + { + "epoch": 1.641487308528571, + "grad_norm": 0.0007352904067374766, + "learning_rate": 0.01858806574758676, + "loss": 0.0843, + "num_input_tokens_seen": 43552832, + "step": 33595 + }, + { + "epoch": 1.6417316101922654, + "grad_norm": 0.0006078292499296367, + "learning_rate": 0.01855967392558988, + "loss": 0.0677, + "num_input_tokens_seen": 43559424, + "step": 33600 + }, + { + "epoch": 1.6417316101922654, + "eval_loss": 0.08458495140075684, + "eval_runtime": 402.66, + "eval_samples_per_second": 90.362, + "eval_steps_per_second": 22.592, + "num_input_tokens_seen": 43559424, + "step": 33600 + }, + { + "epoch": 1.6419759118559596, + "grad_norm": 0.000772194005548954, + "learning_rate": 0.018531302373343096, + "loss": 0.0694, + "num_input_tokens_seen": 43565664, + "step": 33605 + }, + { + "epoch": 1.642220213519654, + "grad_norm": 0.001914251479320228, + "learning_rate": 0.018502951095221588, + "loss": 0.0798, + "num_input_tokens_seen": 43572000, + "step": 33610 + }, + { + "epoch": 1.6424645151833484, + "grad_norm": 0.0009725795825943351, + "learning_rate": 0.01847462009559751, + "loss": 0.0884, + "num_input_tokens_seen": 43578304, + "step": 33615 + }, + { + "epoch": 1.6427088168470427, + "grad_norm": 0.0019437171285972, + "learning_rate": 0.01844630937883992, + "loss": 0.0752, + "num_input_tokens_seen": 43584672, + "step": 33620 + }, + { + "epoch": 1.642953118510737, + "grad_norm": 0.0014477000804618, + "learning_rate": 0.018418018949314573, + "loss": 0.0887, + "num_input_tokens_seen": 43591072, + "step": 33625 + }, + { + "epoch": 1.6431974201744315, + "grad_norm": 0.0014823646051809192, + "learning_rate": 0.018389748811384315, + "loss": 0.1013, + "num_input_tokens_seen": 43597056, + "step": 33630 + }, + { + "epoch": 1.6434417218381259, + "grad_norm": 0.0018611765699461102, + "learning_rate": 0.018361498969408658, + "loss": 0.0963, + "num_input_tokens_seen": 43603584, + "step": 33635 + }, + { + "epoch": 1.64368602350182, + "grad_norm": 0.0026257061399519444, + "learning_rate": 0.01833326942774415, + "loss": 0.1146, + "num_input_tokens_seen": 43609984, + "step": 33640 + }, + { + "epoch": 1.6439303251655144, + "grad_norm": 0.0009715115884318948, + "learning_rate": 0.018305060190744155, + "loss": 0.0689, + "num_input_tokens_seen": 43616544, + "step": 33645 + }, + { + "epoch": 1.6441746268292086, + "grad_norm": 0.00227921805344522, + "learning_rate": 0.018276871262758846, + "loss": 0.0984, + "num_input_tokens_seen": 43622912, + "step": 33650 + }, + { + "epoch": 1.644418928492903, + "grad_norm": 0.0015480491565540433, + "learning_rate": 0.0182487026481353, + "loss": 0.0855, + "num_input_tokens_seen": 43629984, + "step": 33655 + }, + { + "epoch": 1.6446632301565973, + "grad_norm": 0.0011392373126000166, + "learning_rate": 0.018220554351217538, + "loss": 0.0785, + "num_input_tokens_seen": 43636480, + "step": 33660 + }, + { + "epoch": 1.6449075318202917, + "grad_norm": 0.0012763686245307326, + "learning_rate": 0.01819242637634629, + "loss": 0.0647, + "num_input_tokens_seen": 43643008, + "step": 33665 + }, + { + "epoch": 1.645151833483986, + "grad_norm": 0.001832621986977756, + "learning_rate": 0.01816431872785933, + "loss": 0.0967, + "num_input_tokens_seen": 43648928, + "step": 33670 + }, + { + "epoch": 1.6453961351476805, + "grad_norm": 0.0008994376403279603, + "learning_rate": 0.018136231410091148, + "loss": 0.0898, + "num_input_tokens_seen": 43655328, + "step": 33675 + }, + { + "epoch": 1.6456404368113748, + "grad_norm": 0.0013344077160581946, + "learning_rate": 0.018108164427373175, + "loss": 0.0988, + "num_input_tokens_seen": 43661664, + "step": 33680 + }, + { + "epoch": 1.645884738475069, + "grad_norm": 0.0011157143162563443, + "learning_rate": 0.01808011778403375, + "loss": 0.0925, + "num_input_tokens_seen": 43667712, + "step": 33685 + }, + { + "epoch": 1.6461290401387634, + "grad_norm": 0.0009473928948864341, + "learning_rate": 0.01805209148439793, + "loss": 0.0839, + "num_input_tokens_seen": 43674112, + "step": 33690 + }, + { + "epoch": 1.6463733418024575, + "grad_norm": 0.0012860220158472657, + "learning_rate": 0.018024085532787757, + "loss": 0.066, + "num_input_tokens_seen": 43680256, + "step": 33695 + }, + { + "epoch": 1.646617643466152, + "grad_norm": 0.001022863551042974, + "learning_rate": 0.017996099933522164, + "loss": 0.0768, + "num_input_tokens_seen": 43686432, + "step": 33700 + }, + { + "epoch": 1.6468619451298463, + "grad_norm": 0.0016788210486993194, + "learning_rate": 0.017968134690916775, + "loss": 0.0763, + "num_input_tokens_seen": 43692800, + "step": 33705 + }, + { + "epoch": 1.6471062467935407, + "grad_norm": 0.001108501572161913, + "learning_rate": 0.017940189809284263, + "loss": 0.092, + "num_input_tokens_seen": 43699392, + "step": 33710 + }, + { + "epoch": 1.647350548457235, + "grad_norm": 0.0018016460817307234, + "learning_rate": 0.017912265292934024, + "loss": 0.0998, + "num_input_tokens_seen": 43705792, + "step": 33715 + }, + { + "epoch": 1.6475948501209294, + "grad_norm": 0.0012622094945982099, + "learning_rate": 0.017884361146172423, + "loss": 0.0604, + "num_input_tokens_seen": 43712256, + "step": 33720 + }, + { + "epoch": 1.6478391517846238, + "grad_norm": 0.0012911942321807146, + "learning_rate": 0.01785647737330261, + "loss": 0.0983, + "num_input_tokens_seen": 43718912, + "step": 33725 + }, + { + "epoch": 1.648083453448318, + "grad_norm": 0.0007314875838346779, + "learning_rate": 0.017828613978624563, + "loss": 0.0614, + "num_input_tokens_seen": 43725248, + "step": 33730 + }, + { + "epoch": 1.6483277551120123, + "grad_norm": 0.001527767744846642, + "learning_rate": 0.01780077096643523, + "loss": 0.0425, + "num_input_tokens_seen": 43731936, + "step": 33735 + }, + { + "epoch": 1.6485720567757065, + "grad_norm": 0.001490639871917665, + "learning_rate": 0.017772948341028345, + "loss": 0.0843, + "num_input_tokens_seen": 43738208, + "step": 33740 + }, + { + "epoch": 1.6488163584394009, + "grad_norm": 0.002082864288240671, + "learning_rate": 0.01774514610669447, + "loss": 0.0972, + "num_input_tokens_seen": 43744608, + "step": 33745 + }, + { + "epoch": 1.6490606601030953, + "grad_norm": 0.0018811977934092283, + "learning_rate": 0.017717364267721112, + "loss": 0.0779, + "num_input_tokens_seen": 43751072, + "step": 33750 + }, + { + "epoch": 1.6493049617667896, + "grad_norm": 0.0008309705299325287, + "learning_rate": 0.017689602828392513, + "loss": 0.0739, + "num_input_tokens_seen": 43757312, + "step": 33755 + }, + { + "epoch": 1.649549263430484, + "grad_norm": 0.0015664588427171111, + "learning_rate": 0.017661861792989897, + "loss": 0.1023, + "num_input_tokens_seen": 43763616, + "step": 33760 + }, + { + "epoch": 1.6497935650941784, + "grad_norm": 0.001292176079005003, + "learning_rate": 0.017634141165791272, + "loss": 0.0831, + "num_input_tokens_seen": 43770240, + "step": 33765 + }, + { + "epoch": 1.6500378667578728, + "grad_norm": 0.0009437963017262518, + "learning_rate": 0.017606440951071455, + "loss": 0.0552, + "num_input_tokens_seen": 43776672, + "step": 33770 + }, + { + "epoch": 1.650282168421567, + "grad_norm": 0.0018028086051344872, + "learning_rate": 0.017578761153102213, + "loss": 0.1087, + "num_input_tokens_seen": 43782848, + "step": 33775 + }, + { + "epoch": 1.6505264700852613, + "grad_norm": 0.0018503512255847454, + "learning_rate": 0.017551101776152146, + "loss": 0.0844, + "num_input_tokens_seen": 43789376, + "step": 33780 + }, + { + "epoch": 1.6507707717489555, + "grad_norm": 0.0011144017335027456, + "learning_rate": 0.017523462824486608, + "loss": 0.068, + "num_input_tokens_seen": 43795872, + "step": 33785 + }, + { + "epoch": 1.6510150734126499, + "grad_norm": 0.0009535767021588981, + "learning_rate": 0.01749584430236794, + "loss": 0.057, + "num_input_tokens_seen": 43802688, + "step": 33790 + }, + { + "epoch": 1.6512593750763442, + "grad_norm": 0.0018027505138888955, + "learning_rate": 0.01746824621405524, + "loss": 0.0923, + "num_input_tokens_seen": 43809024, + "step": 33795 + }, + { + "epoch": 1.6515036767400386, + "grad_norm": 0.001564512960612774, + "learning_rate": 0.017440668563804412, + "loss": 0.1163, + "num_input_tokens_seen": 43815424, + "step": 33800 + }, + { + "epoch": 1.6515036767400386, + "eval_loss": 0.08438260108232498, + "eval_runtime": 401.9819, + "eval_samples_per_second": 90.514, + "eval_steps_per_second": 22.63, + "num_input_tokens_seen": 43815424, + "step": 33800 + }, + { + "epoch": 1.651747978403733, + "grad_norm": 0.0015999736497178674, + "learning_rate": 0.017413111355868392, + "loss": 0.0892, + "num_input_tokens_seen": 43822720, + "step": 33805 + }, + { + "epoch": 1.6519922800674274, + "grad_norm": 0.0013757452834397554, + "learning_rate": 0.017385574594496748, + "loss": 0.0823, + "num_input_tokens_seen": 43829088, + "step": 33810 + }, + { + "epoch": 1.6522365817311215, + "grad_norm": 0.0016809707740321755, + "learning_rate": 0.01735805828393605, + "loss": 0.0799, + "num_input_tokens_seen": 43835392, + "step": 33815 + }, + { + "epoch": 1.652480883394816, + "grad_norm": 0.0015359153039753437, + "learning_rate": 0.017330562428429667, + "loss": 0.0936, + "num_input_tokens_seen": 43841600, + "step": 33820 + }, + { + "epoch": 1.6527251850585103, + "grad_norm": 0.0016043869545683265, + "learning_rate": 0.01730308703221776, + "loss": 0.0708, + "num_input_tokens_seen": 43848096, + "step": 33825 + }, + { + "epoch": 1.6529694867222045, + "grad_norm": 0.0008392132585868239, + "learning_rate": 0.01727563209953744, + "loss": 0.072, + "num_input_tokens_seen": 43854752, + "step": 33830 + }, + { + "epoch": 1.6532137883858988, + "grad_norm": 0.0008138305856846273, + "learning_rate": 0.017248197634622535, + "loss": 0.0644, + "num_input_tokens_seen": 43861408, + "step": 33835 + }, + { + "epoch": 1.6534580900495932, + "grad_norm": 0.0011241837637498975, + "learning_rate": 0.01722078364170383, + "loss": 0.0631, + "num_input_tokens_seen": 43867936, + "step": 33840 + }, + { + "epoch": 1.6537023917132876, + "grad_norm": 0.001242264756001532, + "learning_rate": 0.017193390125008905, + "loss": 0.0963, + "num_input_tokens_seen": 43874592, + "step": 33845 + }, + { + "epoch": 1.653946693376982, + "grad_norm": 0.0013523829402402043, + "learning_rate": 0.017166017088762153, + "loss": 0.0925, + "num_input_tokens_seen": 43881024, + "step": 33850 + }, + { + "epoch": 1.6541909950406763, + "grad_norm": 0.0012978485319763422, + "learning_rate": 0.017138664537184878, + "loss": 0.0577, + "num_input_tokens_seen": 43887328, + "step": 33855 + }, + { + "epoch": 1.6544352967043705, + "grad_norm": 0.0014269102830439806, + "learning_rate": 0.017111332474495172, + "loss": 0.086, + "num_input_tokens_seen": 43893856, + "step": 33860 + }, + { + "epoch": 1.6546795983680649, + "grad_norm": 0.0019337126286700368, + "learning_rate": 0.017084020904907998, + "loss": 0.083, + "num_input_tokens_seen": 43900352, + "step": 33865 + }, + { + "epoch": 1.6549239000317593, + "grad_norm": 0.0014712372794747353, + "learning_rate": 0.017056729832635103, + "loss": 0.0636, + "num_input_tokens_seen": 43906656, + "step": 33870 + }, + { + "epoch": 1.6551682016954534, + "grad_norm": 0.001298548188060522, + "learning_rate": 0.017029459261885153, + "loss": 0.0734, + "num_input_tokens_seen": 43912960, + "step": 33875 + }, + { + "epoch": 1.6554125033591478, + "grad_norm": 0.0015342364786192775, + "learning_rate": 0.01700220919686359, + "loss": 0.0906, + "num_input_tokens_seen": 43919360, + "step": 33880 + }, + { + "epoch": 1.6556568050228422, + "grad_norm": 0.001532789203338325, + "learning_rate": 0.016974979641772723, + "loss": 0.0774, + "num_input_tokens_seen": 43925696, + "step": 33885 + }, + { + "epoch": 1.6559011066865366, + "grad_norm": 0.0016659105895087123, + "learning_rate": 0.01694777060081169, + "loss": 0.1025, + "num_input_tokens_seen": 43932128, + "step": 33890 + }, + { + "epoch": 1.656145408350231, + "grad_norm": 0.0015579222235828638, + "learning_rate": 0.016920582078176444, + "loss": 0.0835, + "num_input_tokens_seen": 43938368, + "step": 33895 + }, + { + "epoch": 1.6563897100139253, + "grad_norm": 0.0008545920136384666, + "learning_rate": 0.016893414078059863, + "loss": 0.0559, + "num_input_tokens_seen": 43944768, + "step": 33900 + }, + { + "epoch": 1.6566340116776195, + "grad_norm": 0.0012990828836336732, + "learning_rate": 0.016866266604651535, + "loss": 0.0772, + "num_input_tokens_seen": 43951296, + "step": 33905 + }, + { + "epoch": 1.6568783133413139, + "grad_norm": 0.0017465815180912614, + "learning_rate": 0.016839139662137976, + "loss": 0.0778, + "num_input_tokens_seen": 43957952, + "step": 33910 + }, + { + "epoch": 1.6571226150050082, + "grad_norm": 0.0008923541172407568, + "learning_rate": 0.01681203325470245, + "loss": 0.0862, + "num_input_tokens_seen": 43964320, + "step": 33915 + }, + { + "epoch": 1.6573669166687024, + "grad_norm": 0.0017752957064658403, + "learning_rate": 0.016784947386525157, + "loss": 0.0995, + "num_input_tokens_seen": 43970976, + "step": 33920 + }, + { + "epoch": 1.6576112183323968, + "grad_norm": 0.0016428445233032107, + "learning_rate": 0.01675788206178308, + "loss": 0.0597, + "num_input_tokens_seen": 43977472, + "step": 33925 + }, + { + "epoch": 1.6578555199960912, + "grad_norm": 0.0009432046208530664, + "learning_rate": 0.016730837284649986, + "loss": 0.0644, + "num_input_tokens_seen": 43983808, + "step": 33930 + }, + { + "epoch": 1.6580998216597855, + "grad_norm": 0.0016246832674369216, + "learning_rate": 0.016703813059296583, + "loss": 0.1172, + "num_input_tokens_seen": 43989920, + "step": 33935 + }, + { + "epoch": 1.65834412332348, + "grad_norm": 0.0009308931767009199, + "learning_rate": 0.016676809389890294, + "loss": 0.0556, + "num_input_tokens_seen": 43996192, + "step": 33940 + }, + { + "epoch": 1.6585884249871743, + "grad_norm": 0.0014274093555286527, + "learning_rate": 0.016649826280595435, + "loss": 0.0804, + "num_input_tokens_seen": 44002656, + "step": 33945 + }, + { + "epoch": 1.6588327266508684, + "grad_norm": 0.001993870362639427, + "learning_rate": 0.016622863735573163, + "loss": 0.0954, + "num_input_tokens_seen": 44008960, + "step": 33950 + }, + { + "epoch": 1.6590770283145628, + "grad_norm": 0.0014253462431952357, + "learning_rate": 0.016595921758981395, + "loss": 0.0641, + "num_input_tokens_seen": 44015616, + "step": 33955 + }, + { + "epoch": 1.659321329978257, + "grad_norm": 0.0011703235795721412, + "learning_rate": 0.01656900035497495, + "loss": 0.0623, + "num_input_tokens_seen": 44022816, + "step": 33960 + }, + { + "epoch": 1.6595656316419514, + "grad_norm": 0.0013022253988310695, + "learning_rate": 0.016542099527705485, + "loss": 0.0722, + "num_input_tokens_seen": 44029344, + "step": 33965 + }, + { + "epoch": 1.6598099333056457, + "grad_norm": 0.00170897098723799, + "learning_rate": 0.01651521928132138, + "loss": 0.0729, + "num_input_tokens_seen": 44035776, + "step": 33970 + }, + { + "epoch": 1.6600542349693401, + "grad_norm": 0.0008473880589008331, + "learning_rate": 0.01648835961996794, + "loss": 0.1069, + "num_input_tokens_seen": 44041984, + "step": 33975 + }, + { + "epoch": 1.6602985366330345, + "grad_norm": 0.000890266674105078, + "learning_rate": 0.016461520547787285, + "loss": 0.0667, + "num_input_tokens_seen": 44048576, + "step": 33980 + }, + { + "epoch": 1.6605428382967289, + "grad_norm": 0.0012677825288847089, + "learning_rate": 0.016434702068918266, + "loss": 0.0514, + "num_input_tokens_seen": 44054976, + "step": 33985 + }, + { + "epoch": 1.6607871399604233, + "grad_norm": 0.0010115645127370954, + "learning_rate": 0.01640790418749673, + "loss": 0.0728, + "num_input_tokens_seen": 44061280, + "step": 33990 + }, + { + "epoch": 1.6610314416241174, + "grad_norm": 0.002088771667331457, + "learning_rate": 0.016381126907655134, + "loss": 0.0978, + "num_input_tokens_seen": 44067936, + "step": 33995 + }, + { + "epoch": 1.6612757432878118, + "grad_norm": 0.0019760639406740665, + "learning_rate": 0.016354370233522948, + "loss": 0.0684, + "num_input_tokens_seen": 44074432, + "step": 34000 + }, + { + "epoch": 1.6612757432878118, + "eval_loss": 0.08622909337282181, + "eval_runtime": 402.9095, + "eval_samples_per_second": 90.306, + "eval_steps_per_second": 22.578, + "num_input_tokens_seen": 44074432, + "step": 34000 + }, + { + "epoch": 1.661520044951506, + "grad_norm": 0.0011583591112866998, + "learning_rate": 0.016327634169226394, + "loss": 0.0887, + "num_input_tokens_seen": 44081120, + "step": 34005 + }, + { + "epoch": 1.6617643466152003, + "grad_norm": 0.0008779179188422859, + "learning_rate": 0.016300918718888485, + "loss": 0.0668, + "num_input_tokens_seen": 44087424, + "step": 34010 + }, + { + "epoch": 1.6620086482788947, + "grad_norm": 0.0009801653213799, + "learning_rate": 0.016274223886629052, + "loss": 0.0772, + "num_input_tokens_seen": 44093856, + "step": 34015 + }, + { + "epoch": 1.662252949942589, + "grad_norm": 0.0008228968945331872, + "learning_rate": 0.01624754967656482, + "loss": 0.0703, + "num_input_tokens_seen": 44100384, + "step": 34020 + }, + { + "epoch": 1.6624972516062835, + "grad_norm": 0.0015383823774755, + "learning_rate": 0.016220896092809235, + "loss": 0.0815, + "num_input_tokens_seen": 44106752, + "step": 34025 + }, + { + "epoch": 1.6627415532699779, + "grad_norm": 0.0019026939989998937, + "learning_rate": 0.01619426313947267, + "loss": 0.1036, + "num_input_tokens_seen": 44113632, + "step": 34030 + }, + { + "epoch": 1.6629858549336722, + "grad_norm": 0.003004231723025441, + "learning_rate": 0.016167650820662228, + "loss": 0.0779, + "num_input_tokens_seen": 44120096, + "step": 34035 + }, + { + "epoch": 1.6632301565973664, + "grad_norm": 0.0012010844657197595, + "learning_rate": 0.016141059140481855, + "loss": 0.0624, + "num_input_tokens_seen": 44126656, + "step": 34040 + }, + { + "epoch": 1.6634744582610608, + "grad_norm": 0.0014277173904702067, + "learning_rate": 0.016114488103032374, + "loss": 0.067, + "num_input_tokens_seen": 44132896, + "step": 34045 + }, + { + "epoch": 1.663718759924755, + "grad_norm": 0.0016876354347914457, + "learning_rate": 0.016087937712411293, + "loss": 0.0835, + "num_input_tokens_seen": 44139008, + "step": 34050 + }, + { + "epoch": 1.6639630615884493, + "grad_norm": 0.0016164849512279034, + "learning_rate": 0.01606140797271308, + "loss": 0.0866, + "num_input_tokens_seen": 44145728, + "step": 34055 + }, + { + "epoch": 1.6642073632521437, + "grad_norm": 0.0013359084259718657, + "learning_rate": 0.01603489888802897, + "loss": 0.0798, + "num_input_tokens_seen": 44152896, + "step": 34060 + }, + { + "epoch": 1.664451664915838, + "grad_norm": 0.0011403022799640894, + "learning_rate": 0.016008410462446918, + "loss": 0.066, + "num_input_tokens_seen": 44159520, + "step": 34065 + }, + { + "epoch": 1.6646959665795324, + "grad_norm": 0.00164665631018579, + "learning_rate": 0.01598194270005185, + "loss": 0.0859, + "num_input_tokens_seen": 44166048, + "step": 34070 + }, + { + "epoch": 1.6649402682432268, + "grad_norm": 0.0014350088313221931, + "learning_rate": 0.015955495604925356, + "loss": 0.0741, + "num_input_tokens_seen": 44172288, + "step": 34075 + }, + { + "epoch": 1.6651845699069212, + "grad_norm": 0.0009129217942245305, + "learning_rate": 0.01592906918114598, + "loss": 0.0801, + "num_input_tokens_seen": 44178464, + "step": 34080 + }, + { + "epoch": 1.6654288715706154, + "grad_norm": 0.001490610302425921, + "learning_rate": 0.015902663432788965, + "loss": 0.0941, + "num_input_tokens_seen": 44184928, + "step": 34085 + }, + { + "epoch": 1.6656731732343097, + "grad_norm": 0.0012604092480614781, + "learning_rate": 0.01587627836392643, + "loss": 0.0846, + "num_input_tokens_seen": 44191456, + "step": 34090 + }, + { + "epoch": 1.665917474898004, + "grad_norm": 0.001390933757647872, + "learning_rate": 0.01584991397862726, + "loss": 0.0802, + "num_input_tokens_seen": 44198080, + "step": 34095 + }, + { + "epoch": 1.6661617765616983, + "grad_norm": 0.0016692880308255553, + "learning_rate": 0.015823570280957214, + "loss": 0.0949, + "num_input_tokens_seen": 44204672, + "step": 34100 + }, + { + "epoch": 1.6664060782253927, + "grad_norm": 0.0008936035446822643, + "learning_rate": 0.015797247274978766, + "loss": 0.0714, + "num_input_tokens_seen": 44211040, + "step": 34105 + }, + { + "epoch": 1.666650379889087, + "grad_norm": 0.0007895657909102738, + "learning_rate": 0.015770944964751326, + "loss": 0.0761, + "num_input_tokens_seen": 44217728, + "step": 34110 + }, + { + "epoch": 1.6668946815527814, + "grad_norm": 0.001619074959307909, + "learning_rate": 0.015744663354330956, + "loss": 0.0756, + "num_input_tokens_seen": 44224000, + "step": 34115 + }, + { + "epoch": 1.6671389832164758, + "grad_norm": 0.0016335368854925036, + "learning_rate": 0.015718402447770664, + "loss": 0.0695, + "num_input_tokens_seen": 44230752, + "step": 34120 + }, + { + "epoch": 1.6673832848801702, + "grad_norm": 0.001725290436297655, + "learning_rate": 0.015692162249120224, + "loss": 0.076, + "num_input_tokens_seen": 44237312, + "step": 34125 + }, + { + "epoch": 1.6676275865438643, + "grad_norm": 0.0017251316457986832, + "learning_rate": 0.01566594276242615, + "loss": 0.0981, + "num_input_tokens_seen": 44243712, + "step": 34130 + }, + { + "epoch": 1.6678718882075587, + "grad_norm": 0.0013886078959330916, + "learning_rate": 0.015639743991731857, + "loss": 0.1056, + "num_input_tokens_seen": 44249856, + "step": 34135 + }, + { + "epoch": 1.6681161898712529, + "grad_norm": 0.0009203300578519702, + "learning_rate": 0.01561356594107755, + "loss": 0.0801, + "num_input_tokens_seen": 44256160, + "step": 34140 + }, + { + "epoch": 1.6683604915349473, + "grad_norm": 0.0019175385823473334, + "learning_rate": 0.015587408614500147, + "loss": 0.1006, + "num_input_tokens_seen": 44262400, + "step": 34145 + }, + { + "epoch": 1.6686047931986416, + "grad_norm": 0.001960858004167676, + "learning_rate": 0.015561272016033505, + "loss": 0.0922, + "num_input_tokens_seen": 44268800, + "step": 34150 + }, + { + "epoch": 1.668849094862336, + "grad_norm": 0.0011985511519014835, + "learning_rate": 0.015535156149708167, + "loss": 0.0929, + "num_input_tokens_seen": 44274784, + "step": 34155 + }, + { + "epoch": 1.6690933965260304, + "grad_norm": 0.0013256918173283339, + "learning_rate": 0.015509061019551528, + "loss": 0.1, + "num_input_tokens_seen": 44281152, + "step": 34160 + }, + { + "epoch": 1.6693376981897248, + "grad_norm": 0.0015113485278561711, + "learning_rate": 0.015482986629587818, + "loss": 0.0675, + "num_input_tokens_seen": 44287904, + "step": 34165 + }, + { + "epoch": 1.6695819998534192, + "grad_norm": 0.0010203178972005844, + "learning_rate": 0.01545693298383799, + "loss": 0.0692, + "num_input_tokens_seen": 44294496, + "step": 34170 + }, + { + "epoch": 1.6698263015171133, + "grad_norm": 0.0018505214247852564, + "learning_rate": 0.015430900086319858, + "loss": 0.0841, + "num_input_tokens_seen": 44301120, + "step": 34175 + }, + { + "epoch": 1.6700706031808077, + "grad_norm": 0.0016169368755072355, + "learning_rate": 0.015404887941048084, + "loss": 0.0853, + "num_input_tokens_seen": 44308096, + "step": 34180 + }, + { + "epoch": 1.6703149048445018, + "grad_norm": 0.001687323790974915, + "learning_rate": 0.01537889655203397, + "loss": 0.0957, + "num_input_tokens_seen": 44314592, + "step": 34185 + }, + { + "epoch": 1.6705592065081962, + "grad_norm": 0.0012427830370143056, + "learning_rate": 0.015352925923285798, + "loss": 0.0674, + "num_input_tokens_seen": 44321152, + "step": 34190 + }, + { + "epoch": 1.6708035081718906, + "grad_norm": 0.0008485346334055066, + "learning_rate": 0.015326976058808511, + "loss": 0.1062, + "num_input_tokens_seen": 44327584, + "step": 34195 + }, + { + "epoch": 1.671047809835585, + "grad_norm": 0.0016917820321395993, + "learning_rate": 0.015301046962603908, + "loss": 0.0725, + "num_input_tokens_seen": 44334304, + "step": 34200 + }, + { + "epoch": 1.671047809835585, + "eval_loss": 0.08467563986778259, + "eval_runtime": 403.0608, + "eval_samples_per_second": 90.272, + "eval_steps_per_second": 22.57, + "num_input_tokens_seen": 44334304, + "step": 34200 + }, + { + "epoch": 1.6712921114992794, + "grad_norm": 0.0011667495127767324, + "learning_rate": 0.015275138638670626, + "loss": 0.0736, + "num_input_tokens_seen": 44340928, + "step": 34205 + }, + { + "epoch": 1.6715364131629737, + "grad_norm": 0.0008867579163052142, + "learning_rate": 0.015249251091004001, + "loss": 0.078, + "num_input_tokens_seen": 44347616, + "step": 34210 + }, + { + "epoch": 1.6717807148266681, + "grad_norm": 0.0018558554584160447, + "learning_rate": 0.01522338432359624, + "loss": 0.0849, + "num_input_tokens_seen": 44354240, + "step": 34215 + }, + { + "epoch": 1.6720250164903623, + "grad_norm": 0.0012947374489158392, + "learning_rate": 0.01519753834043635, + "loss": 0.0867, + "num_input_tokens_seen": 44361024, + "step": 34220 + }, + { + "epoch": 1.6722693181540567, + "grad_norm": 0.0009625963866710663, + "learning_rate": 0.015171713145510095, + "loss": 0.0747, + "num_input_tokens_seen": 44367232, + "step": 34225 + }, + { + "epoch": 1.6725136198177508, + "grad_norm": 0.0013031007256358862, + "learning_rate": 0.01514590874279999, + "loss": 0.0969, + "num_input_tokens_seen": 44374176, + "step": 34230 + }, + { + "epoch": 1.6727579214814452, + "grad_norm": 0.0008098417893052101, + "learning_rate": 0.015120125136285467, + "loss": 0.0743, + "num_input_tokens_seen": 44380416, + "step": 34235 + }, + { + "epoch": 1.6730022231451396, + "grad_norm": 0.0011778941843658686, + "learning_rate": 0.015094362329942629, + "loss": 0.0539, + "num_input_tokens_seen": 44386720, + "step": 34240 + }, + { + "epoch": 1.673246524808834, + "grad_norm": 0.0011738208122551441, + "learning_rate": 0.01506862032774448, + "loss": 0.0634, + "num_input_tokens_seen": 44393152, + "step": 34245 + }, + { + "epoch": 1.6734908264725283, + "grad_norm": 0.0011392877204343677, + "learning_rate": 0.015042899133660697, + "loss": 0.0808, + "num_input_tokens_seen": 44399808, + "step": 34250 + }, + { + "epoch": 1.6737351281362227, + "grad_norm": 0.0009648574632592499, + "learning_rate": 0.01501719875165789, + "loss": 0.0813, + "num_input_tokens_seen": 44406336, + "step": 34255 + }, + { + "epoch": 1.673979429799917, + "grad_norm": 0.0011823943350464106, + "learning_rate": 0.014991519185699286, + "loss": 0.0742, + "num_input_tokens_seen": 44413312, + "step": 34260 + }, + { + "epoch": 1.6742237314636113, + "grad_norm": 0.0012287880526855588, + "learning_rate": 0.014965860439745054, + "loss": 0.0628, + "num_input_tokens_seen": 44419488, + "step": 34265 + }, + { + "epoch": 1.6744680331273056, + "grad_norm": 0.0015620924532413483, + "learning_rate": 0.01494022251775211, + "loss": 0.0864, + "num_input_tokens_seen": 44426048, + "step": 34270 + }, + { + "epoch": 1.6747123347909998, + "grad_norm": 0.0018094762926921248, + "learning_rate": 0.014914605423674109, + "loss": 0.0961, + "num_input_tokens_seen": 44432576, + "step": 34275 + }, + { + "epoch": 1.6749566364546942, + "grad_norm": 0.0023779638577252626, + "learning_rate": 0.014889009161461525, + "loss": 0.071, + "num_input_tokens_seen": 44439296, + "step": 34280 + }, + { + "epoch": 1.6752009381183885, + "grad_norm": 0.001127955736592412, + "learning_rate": 0.014863433735061665, + "loss": 0.0874, + "num_input_tokens_seen": 44445568, + "step": 34285 + }, + { + "epoch": 1.675445239782083, + "grad_norm": 0.0011044185375794768, + "learning_rate": 0.014837879148418541, + "loss": 0.0661, + "num_input_tokens_seen": 44452288, + "step": 34290 + }, + { + "epoch": 1.6756895414457773, + "grad_norm": 0.0011114644585177302, + "learning_rate": 0.01481234540547302, + "loss": 0.0677, + "num_input_tokens_seen": 44458848, + "step": 34295 + }, + { + "epoch": 1.6759338431094717, + "grad_norm": 0.001520649646408856, + "learning_rate": 0.014786832510162717, + "loss": 0.0847, + "num_input_tokens_seen": 44465472, + "step": 34300 + }, + { + "epoch": 1.676178144773166, + "grad_norm": 0.002225398551672697, + "learning_rate": 0.014761340466422017, + "loss": 0.1009, + "num_input_tokens_seen": 44471840, + "step": 34305 + }, + { + "epoch": 1.6764224464368602, + "grad_norm": 0.0014719241298735142, + "learning_rate": 0.014735869278182144, + "loss": 0.088, + "num_input_tokens_seen": 44478112, + "step": 34310 + }, + { + "epoch": 1.6766667481005546, + "grad_norm": 0.0013066122774034739, + "learning_rate": 0.014710418949371057, + "loss": 0.0716, + "num_input_tokens_seen": 44484832, + "step": 34315 + }, + { + "epoch": 1.6769110497642488, + "grad_norm": 0.0009315394563600421, + "learning_rate": 0.014684989483913495, + "loss": 0.068, + "num_input_tokens_seen": 44491744, + "step": 34320 + }, + { + "epoch": 1.6771553514279431, + "grad_norm": 0.0008475275244563818, + "learning_rate": 0.014659580885731077, + "loss": 0.0759, + "num_input_tokens_seen": 44498336, + "step": 34325 + }, + { + "epoch": 1.6773996530916375, + "grad_norm": 0.0008577378466725349, + "learning_rate": 0.014634193158742047, + "loss": 0.081, + "num_input_tokens_seen": 44504672, + "step": 34330 + }, + { + "epoch": 1.677643954755332, + "grad_norm": 0.0013777506537735462, + "learning_rate": 0.014608826306861576, + "loss": 0.0748, + "num_input_tokens_seen": 44511488, + "step": 34335 + }, + { + "epoch": 1.6778882564190263, + "grad_norm": 0.0016444643260911107, + "learning_rate": 0.014583480334001486, + "loss": 0.0643, + "num_input_tokens_seen": 44517760, + "step": 34340 + }, + { + "epoch": 1.6781325580827207, + "grad_norm": 0.0014593895757570863, + "learning_rate": 0.014558155244070496, + "loss": 0.0891, + "num_input_tokens_seen": 44524512, + "step": 34345 + }, + { + "epoch": 1.678376859746415, + "grad_norm": 0.0019293330842629075, + "learning_rate": 0.014532851040974036, + "loss": 0.0778, + "num_input_tokens_seen": 44530560, + "step": 34350 + }, + { + "epoch": 1.6786211614101092, + "grad_norm": 0.0013863217318430543, + "learning_rate": 0.014507567728614335, + "loss": 0.1009, + "num_input_tokens_seen": 44536864, + "step": 34355 + }, + { + "epoch": 1.6788654630738036, + "grad_norm": 0.0016199360834434628, + "learning_rate": 0.01448230531089037, + "loss": 0.0992, + "num_input_tokens_seen": 44543136, + "step": 34360 + }, + { + "epoch": 1.6791097647374977, + "grad_norm": 0.0009989379905164242, + "learning_rate": 0.014457063791697993, + "loss": 0.0534, + "num_input_tokens_seen": 44549984, + "step": 34365 + }, + { + "epoch": 1.6793540664011921, + "grad_norm": 0.0009956822032108903, + "learning_rate": 0.01443184317492971, + "loss": 0.0858, + "num_input_tokens_seen": 44556768, + "step": 34370 + }, + { + "epoch": 1.6795983680648865, + "grad_norm": 0.0006649269489571452, + "learning_rate": 0.014406643464474822, + "loss": 0.0687, + "num_input_tokens_seen": 44562848, + "step": 34375 + }, + { + "epoch": 1.6798426697285809, + "grad_norm": 0.001714316662400961, + "learning_rate": 0.014381464664219539, + "loss": 0.0824, + "num_input_tokens_seen": 44569504, + "step": 34380 + }, + { + "epoch": 1.6800869713922753, + "grad_norm": 0.0009884354658424854, + "learning_rate": 0.014356306778046656, + "loss": 0.1155, + "num_input_tokens_seen": 44575552, + "step": 34385 + }, + { + "epoch": 1.6803312730559696, + "grad_norm": 0.0021651778370141983, + "learning_rate": 0.014331169809835885, + "loss": 0.0698, + "num_input_tokens_seen": 44581856, + "step": 34390 + }, + { + "epoch": 1.6805755747196638, + "grad_norm": 0.0013064637314528227, + "learning_rate": 0.014306053763463644, + "loss": 0.075, + "num_input_tokens_seen": 44588128, + "step": 34395 + }, + { + "epoch": 1.6808198763833582, + "grad_norm": 0.001855981769040227, + "learning_rate": 0.014280958642803147, + "loss": 0.0811, + "num_input_tokens_seen": 44594368, + "step": 34400 + }, + { + "epoch": 1.6808198763833582, + "eval_loss": 0.08430863171815872, + "eval_runtime": 402.2809, + "eval_samples_per_second": 90.447, + "eval_steps_per_second": 22.614, + "num_input_tokens_seen": 44594368, + "step": 34400 + }, + { + "epoch": 1.6810641780470525, + "grad_norm": 0.0022943324875086546, + "learning_rate": 0.014255884451724404, + "loss": 0.1114, + "num_input_tokens_seen": 44600704, + "step": 34405 + }, + { + "epoch": 1.6813084797107467, + "grad_norm": 0.0011047907173633575, + "learning_rate": 0.014230831194094101, + "loss": 0.0578, + "num_input_tokens_seen": 44607392, + "step": 34410 + }, + { + "epoch": 1.681552781374441, + "grad_norm": 0.0010603908449411392, + "learning_rate": 0.014205798873775865, + "loss": 0.0913, + "num_input_tokens_seen": 44613472, + "step": 34415 + }, + { + "epoch": 1.6817970830381355, + "grad_norm": 0.0007527807028964162, + "learning_rate": 0.014180787494629893, + "loss": 0.0918, + "num_input_tokens_seen": 44620096, + "step": 34420 + }, + { + "epoch": 1.6820413847018298, + "grad_norm": 0.0008245244389399886, + "learning_rate": 0.014155797060513314, + "loss": 0.0757, + "num_input_tokens_seen": 44626656, + "step": 34425 + }, + { + "epoch": 1.6822856863655242, + "grad_norm": 0.000877891608979553, + "learning_rate": 0.014130827575279963, + "loss": 0.0879, + "num_input_tokens_seen": 44633152, + "step": 34430 + }, + { + "epoch": 1.6825299880292186, + "grad_norm": 0.0014597749104723334, + "learning_rate": 0.014105879042780427, + "loss": 0.0604, + "num_input_tokens_seen": 44639584, + "step": 34435 + }, + { + "epoch": 1.6827742896929128, + "grad_norm": 0.0009559053578414023, + "learning_rate": 0.014080951466862113, + "loss": 0.0823, + "num_input_tokens_seen": 44645952, + "step": 34440 + }, + { + "epoch": 1.6830185913566071, + "grad_norm": 0.0019458195893093944, + "learning_rate": 0.014056044851369126, + "loss": 0.0882, + "num_input_tokens_seen": 44652288, + "step": 34445 + }, + { + "epoch": 1.6832628930203015, + "grad_norm": 0.0015818976098671556, + "learning_rate": 0.014031159200142428, + "loss": 0.1072, + "num_input_tokens_seen": 44658528, + "step": 34450 + }, + { + "epoch": 1.6835071946839957, + "grad_norm": 0.0016094687161967158, + "learning_rate": 0.014006294517019667, + "loss": 0.0713, + "num_input_tokens_seen": 44665440, + "step": 34455 + }, + { + "epoch": 1.68375149634769, + "grad_norm": 0.001310624647885561, + "learning_rate": 0.013981450805835276, + "loss": 0.0712, + "num_input_tokens_seen": 44672480, + "step": 34460 + }, + { + "epoch": 1.6839957980113844, + "grad_norm": 0.0006974919815547764, + "learning_rate": 0.01395662807042049, + "loss": 0.0691, + "num_input_tokens_seen": 44679008, + "step": 34465 + }, + { + "epoch": 1.6842400996750788, + "grad_norm": 0.0009151027770712972, + "learning_rate": 0.013931826314603296, + "loss": 0.0607, + "num_input_tokens_seen": 44685440, + "step": 34470 + }, + { + "epoch": 1.6844844013387732, + "grad_norm": 0.0023344988003373146, + "learning_rate": 0.013907045542208401, + "loss": 0.0934, + "num_input_tokens_seen": 44691904, + "step": 34475 + }, + { + "epoch": 1.6847287030024676, + "grad_norm": 0.001656251261010766, + "learning_rate": 0.013882285757057333, + "loss": 0.0802, + "num_input_tokens_seen": 44698080, + "step": 34480 + }, + { + "epoch": 1.6849730046661617, + "grad_norm": 0.0013921428471803665, + "learning_rate": 0.013857546962968403, + "loss": 0.0854, + "num_input_tokens_seen": 44704384, + "step": 34485 + }, + { + "epoch": 1.6852173063298561, + "grad_norm": 0.0012170823756605387, + "learning_rate": 0.013832829163756577, + "loss": 0.0937, + "num_input_tokens_seen": 44710688, + "step": 34490 + }, + { + "epoch": 1.6854616079935503, + "grad_norm": 0.0013441360788419843, + "learning_rate": 0.013808132363233689, + "loss": 0.0668, + "num_input_tokens_seen": 44716992, + "step": 34495 + }, + { + "epoch": 1.6857059096572446, + "grad_norm": 0.0019031859701499343, + "learning_rate": 0.013783456565208256, + "loss": 0.1048, + "num_input_tokens_seen": 44723584, + "step": 34500 + }, + { + "epoch": 1.685950211320939, + "grad_norm": 0.0013695730594918132, + "learning_rate": 0.01375880177348564, + "loss": 0.1063, + "num_input_tokens_seen": 44730176, + "step": 34505 + }, + { + "epoch": 1.6861945129846334, + "grad_norm": 0.0036094444803893566, + "learning_rate": 0.013734167991867928, + "loss": 0.0966, + "num_input_tokens_seen": 44736704, + "step": 34510 + }, + { + "epoch": 1.6864388146483278, + "grad_norm": 0.0010957795893773437, + "learning_rate": 0.013709555224153935, + "loss": 0.0967, + "num_input_tokens_seen": 44743136, + "step": 34515 + }, + { + "epoch": 1.6866831163120222, + "grad_norm": 0.0008779058116488159, + "learning_rate": 0.013684963474139222, + "loss": 0.0849, + "num_input_tokens_seen": 44749568, + "step": 34520 + }, + { + "epoch": 1.6869274179757165, + "grad_norm": 0.0013801208697259426, + "learning_rate": 0.013660392745616224, + "loss": 0.0724, + "num_input_tokens_seen": 44756320, + "step": 34525 + }, + { + "epoch": 1.6871717196394107, + "grad_norm": 0.0015360829420387745, + "learning_rate": 0.013635843042373974, + "loss": 0.0747, + "num_input_tokens_seen": 44762208, + "step": 34530 + }, + { + "epoch": 1.687416021303105, + "grad_norm": 0.001395246828906238, + "learning_rate": 0.01361131436819843, + "loss": 0.1003, + "num_input_tokens_seen": 44768192, + "step": 34535 + }, + { + "epoch": 1.6876603229667992, + "grad_norm": 0.0019073780858889222, + "learning_rate": 0.013586806726872147, + "loss": 0.0807, + "num_input_tokens_seen": 44774240, + "step": 34540 + }, + { + "epoch": 1.6879046246304936, + "grad_norm": 0.0014680830063298345, + "learning_rate": 0.013562320122174537, + "loss": 0.068, + "num_input_tokens_seen": 44780672, + "step": 34545 + }, + { + "epoch": 1.688148926294188, + "grad_norm": 0.001403955277055502, + "learning_rate": 0.013537854557881762, + "loss": 0.0959, + "num_input_tokens_seen": 44787104, + "step": 34550 + }, + { + "epoch": 1.6883932279578824, + "grad_norm": 0.0010610767640173435, + "learning_rate": 0.013513410037766687, + "loss": 0.0868, + "num_input_tokens_seen": 44793568, + "step": 34555 + }, + { + "epoch": 1.6886375296215768, + "grad_norm": 0.0012857853434979916, + "learning_rate": 0.013488986565598998, + "loss": 0.106, + "num_input_tokens_seen": 44800224, + "step": 34560 + }, + { + "epoch": 1.6888818312852711, + "grad_norm": 0.0020971307530999184, + "learning_rate": 0.013464584145145097, + "loss": 0.0826, + "num_input_tokens_seen": 44806688, + "step": 34565 + }, + { + "epoch": 1.6891261329489655, + "grad_norm": 0.0008613663958385587, + "learning_rate": 0.013440202780168109, + "loss": 0.0752, + "num_input_tokens_seen": 44813888, + "step": 34570 + }, + { + "epoch": 1.6893704346126597, + "grad_norm": 0.0019344855099916458, + "learning_rate": 0.01341584247442799, + "loss": 0.0694, + "num_input_tokens_seen": 44820832, + "step": 34575 + }, + { + "epoch": 1.689614736276354, + "grad_norm": 0.0014505860162898898, + "learning_rate": 0.013391503231681355, + "loss": 0.1035, + "num_input_tokens_seen": 44826848, + "step": 34580 + }, + { + "epoch": 1.6898590379400482, + "grad_norm": 0.001301547046750784, + "learning_rate": 0.013367185055681685, + "loss": 0.0737, + "num_input_tokens_seen": 44833152, + "step": 34585 + }, + { + "epoch": 1.6901033396037426, + "grad_norm": 0.0013284786837175488, + "learning_rate": 0.013342887950179095, + "loss": 0.0644, + "num_input_tokens_seen": 44839808, + "step": 34590 + }, + { + "epoch": 1.690347641267437, + "grad_norm": 0.00188105134293437, + "learning_rate": 0.013318611918920554, + "loss": 0.0833, + "num_input_tokens_seen": 44846048, + "step": 34595 + }, + { + "epoch": 1.6905919429311314, + "grad_norm": 0.0013822474284097552, + "learning_rate": 0.01329435696564965, + "loss": 0.0729, + "num_input_tokens_seen": 44852576, + "step": 34600 + }, + { + "epoch": 1.6905919429311314, + "eval_loss": 0.08404368907213211, + "eval_runtime": 401.8699, + "eval_samples_per_second": 90.539, + "eval_steps_per_second": 22.637, + "num_input_tokens_seen": 44852576, + "step": 34600 + }, + { + "epoch": 1.6908362445948257, + "grad_norm": 0.0015092737739905715, + "learning_rate": 0.013270123094106894, + "loss": 0.084, + "num_input_tokens_seen": 44858912, + "step": 34605 + }, + { + "epoch": 1.6910805462585201, + "grad_norm": 0.0016053388826549053, + "learning_rate": 0.013245910308029395, + "loss": 0.0789, + "num_input_tokens_seen": 44865248, + "step": 34610 + }, + { + "epoch": 1.6913248479222145, + "grad_norm": 0.0014029907761141658, + "learning_rate": 0.0132217186111511, + "loss": 0.0616, + "num_input_tokens_seen": 44871616, + "step": 34615 + }, + { + "epoch": 1.6915691495859086, + "grad_norm": 0.0018644756637513638, + "learning_rate": 0.013197548007202626, + "loss": 0.0778, + "num_input_tokens_seen": 44878176, + "step": 34620 + }, + { + "epoch": 1.691813451249603, + "grad_norm": 0.0016358280554413795, + "learning_rate": 0.01317339849991142, + "loss": 0.1015, + "num_input_tokens_seen": 44884320, + "step": 34625 + }, + { + "epoch": 1.6920577529132972, + "grad_norm": 0.001214281772263348, + "learning_rate": 0.013149270093001675, + "loss": 0.0871, + "num_input_tokens_seen": 44890464, + "step": 34630 + }, + { + "epoch": 1.6923020545769916, + "grad_norm": 0.0024755552876740694, + "learning_rate": 0.013125162790194227, + "loss": 0.1102, + "num_input_tokens_seen": 44897216, + "step": 34635 + }, + { + "epoch": 1.692546356240686, + "grad_norm": 0.0012353661004453897, + "learning_rate": 0.01310107659520674, + "loss": 0.0619, + "num_input_tokens_seen": 44903744, + "step": 34640 + }, + { + "epoch": 1.6927906579043803, + "grad_norm": 0.0009557275334373116, + "learning_rate": 0.013077011511753655, + "loss": 0.0898, + "num_input_tokens_seen": 44910272, + "step": 34645 + }, + { + "epoch": 1.6930349595680747, + "grad_norm": 0.0010017210152000189, + "learning_rate": 0.013052967543546056, + "loss": 0.0704, + "num_input_tokens_seen": 44916224, + "step": 34650 + }, + { + "epoch": 1.693279261231769, + "grad_norm": 0.0017036956269294024, + "learning_rate": 0.01302894469429186, + "loss": 0.0808, + "num_input_tokens_seen": 44922752, + "step": 34655 + }, + { + "epoch": 1.6935235628954635, + "grad_norm": 0.002642898354679346, + "learning_rate": 0.013004942967695653, + "loss": 0.0905, + "num_input_tokens_seen": 44929568, + "step": 34660 + }, + { + "epoch": 1.6937678645591576, + "grad_norm": 0.0020693400874733925, + "learning_rate": 0.012980962367458859, + "loss": 0.0813, + "num_input_tokens_seen": 44935776, + "step": 34665 + }, + { + "epoch": 1.694012166222852, + "grad_norm": 0.0013064080849289894, + "learning_rate": 0.012957002897279567, + "loss": 0.0924, + "num_input_tokens_seen": 44941632, + "step": 34670 + }, + { + "epoch": 1.6942564678865462, + "grad_norm": 0.0010295406682416797, + "learning_rate": 0.012933064560852576, + "loss": 0.0981, + "num_input_tokens_seen": 44947904, + "step": 34675 + }, + { + "epoch": 1.6945007695502405, + "grad_norm": 0.0009841583669185638, + "learning_rate": 0.012909147361869527, + "loss": 0.0805, + "num_input_tokens_seen": 44953984, + "step": 34680 + }, + { + "epoch": 1.694745071213935, + "grad_norm": 0.0011063116835430264, + "learning_rate": 0.012885251304018774, + "loss": 0.0676, + "num_input_tokens_seen": 44960640, + "step": 34685 + }, + { + "epoch": 1.6949893728776293, + "grad_norm": 0.0015052843373268843, + "learning_rate": 0.012861376390985335, + "loss": 0.0827, + "num_input_tokens_seen": 44967296, + "step": 34690 + }, + { + "epoch": 1.6952336745413237, + "grad_norm": 0.0019084070809185505, + "learning_rate": 0.012837522626451063, + "loss": 0.0686, + "num_input_tokens_seen": 44973728, + "step": 34695 + }, + { + "epoch": 1.695477976205018, + "grad_norm": 0.0008494087378494442, + "learning_rate": 0.01281369001409447, + "loss": 0.0716, + "num_input_tokens_seen": 44980160, + "step": 34700 + }, + { + "epoch": 1.6957222778687124, + "grad_norm": 0.0009649499552324414, + "learning_rate": 0.012789878557590877, + "loss": 0.0778, + "num_input_tokens_seen": 44986560, + "step": 34705 + }, + { + "epoch": 1.6959665795324066, + "grad_norm": 0.0007385750650428236, + "learning_rate": 0.012766088260612334, + "loss": 0.0722, + "num_input_tokens_seen": 44993152, + "step": 34710 + }, + { + "epoch": 1.696210881196101, + "grad_norm": 0.0016481464263051748, + "learning_rate": 0.012742319126827523, + "loss": 0.0773, + "num_input_tokens_seen": 44999328, + "step": 34715 + }, + { + "epoch": 1.6964551828597951, + "grad_norm": 0.0013939343625679612, + "learning_rate": 0.012718571159902008, + "loss": 0.0894, + "num_input_tokens_seen": 45005952, + "step": 34720 + }, + { + "epoch": 1.6966994845234895, + "grad_norm": 0.0015881964936852455, + "learning_rate": 0.01269484436349803, + "loss": 0.0791, + "num_input_tokens_seen": 45012480, + "step": 34725 + }, + { + "epoch": 1.696943786187184, + "grad_norm": 0.0011431104503571987, + "learning_rate": 0.012671138741274528, + "loss": 0.0754, + "num_input_tokens_seen": 45019104, + "step": 34730 + }, + { + "epoch": 1.6971880878508783, + "grad_norm": 0.0010865081567317247, + "learning_rate": 0.012647454296887194, + "loss": 0.0561, + "num_input_tokens_seen": 45025952, + "step": 34735 + }, + { + "epoch": 1.6974323895145726, + "grad_norm": 0.0014602761948481202, + "learning_rate": 0.012623791033988507, + "loss": 0.0842, + "num_input_tokens_seen": 45032448, + "step": 34740 + }, + { + "epoch": 1.697676691178267, + "grad_norm": 0.0015041828155517578, + "learning_rate": 0.012600148956227597, + "loss": 0.0942, + "num_input_tokens_seen": 45038720, + "step": 34745 + }, + { + "epoch": 1.6979209928419614, + "grad_norm": 0.0008823435055091977, + "learning_rate": 0.012576528067250414, + "loss": 0.064, + "num_input_tokens_seen": 45044736, + "step": 34750 + }, + { + "epoch": 1.6981652945056556, + "grad_norm": 0.0010417216690257192, + "learning_rate": 0.012552928370699561, + "loss": 0.0793, + "num_input_tokens_seen": 45051008, + "step": 34755 + }, + { + "epoch": 1.69840959616935, + "grad_norm": 0.0015781373949721456, + "learning_rate": 0.012529349870214411, + "loss": 0.1028, + "num_input_tokens_seen": 45057728, + "step": 34760 + }, + { + "epoch": 1.698653897833044, + "grad_norm": 0.0008539354894310236, + "learning_rate": 0.012505792569431106, + "loss": 0.0839, + "num_input_tokens_seen": 45064160, + "step": 34765 + }, + { + "epoch": 1.6988981994967385, + "grad_norm": 0.001744526787661016, + "learning_rate": 0.012482256471982422, + "loss": 0.0784, + "num_input_tokens_seen": 45070592, + "step": 34770 + }, + { + "epoch": 1.6991425011604329, + "grad_norm": 0.0012079813750460744, + "learning_rate": 0.012458741581497956, + "loss": 0.0663, + "num_input_tokens_seen": 45076864, + "step": 34775 + }, + { + "epoch": 1.6993868028241272, + "grad_norm": 0.0019888661336153746, + "learning_rate": 0.012435247901603974, + "loss": 0.0938, + "num_input_tokens_seen": 45083232, + "step": 34780 + }, + { + "epoch": 1.6996311044878216, + "grad_norm": 0.0015195327578112483, + "learning_rate": 0.012411775435923528, + "loss": 0.1037, + "num_input_tokens_seen": 45089760, + "step": 34785 + }, + { + "epoch": 1.699875406151516, + "grad_norm": 0.0011516864178702235, + "learning_rate": 0.012388324188076354, + "loss": 0.1024, + "num_input_tokens_seen": 45095744, + "step": 34790 + }, + { + "epoch": 1.7001197078152104, + "grad_norm": 0.0012049805372953415, + "learning_rate": 0.012364894161678913, + "loss": 0.0812, + "num_input_tokens_seen": 45102240, + "step": 34795 + }, + { + "epoch": 1.7003640094789045, + "grad_norm": 0.0008802792872302234, + "learning_rate": 0.012341485360344445, + "loss": 0.0726, + "num_input_tokens_seen": 45109056, + "step": 34800 + }, + { + "epoch": 1.7003640094789045, + "eval_loss": 0.08427605777978897, + "eval_runtime": 402.4322, + "eval_samples_per_second": 90.413, + "eval_steps_per_second": 22.605, + "num_input_tokens_seen": 45109056, + "step": 34800 + }, + { + "epoch": 1.700608311142599, + "grad_norm": 0.0006050647352822125, + "learning_rate": 0.01231809778768283, + "loss": 0.0598, + "num_input_tokens_seen": 45115680, + "step": 34805 + }, + { + "epoch": 1.700852612806293, + "grad_norm": 0.0013631823239848018, + "learning_rate": 0.012294731447300799, + "loss": 0.089, + "num_input_tokens_seen": 45122016, + "step": 34810 + }, + { + "epoch": 1.7010969144699875, + "grad_norm": 0.0014477098593488336, + "learning_rate": 0.012271386342801671, + "loss": 0.0799, + "num_input_tokens_seen": 45128512, + "step": 34815 + }, + { + "epoch": 1.7013412161336818, + "grad_norm": 0.0012733289040625095, + "learning_rate": 0.012248062477785565, + "loss": 0.0881, + "num_input_tokens_seen": 45134720, + "step": 34820 + }, + { + "epoch": 1.7015855177973762, + "grad_norm": 0.0017015761695802212, + "learning_rate": 0.012224759855849305, + "loss": 0.101, + "num_input_tokens_seen": 45141120, + "step": 34825 + }, + { + "epoch": 1.7018298194610706, + "grad_norm": 0.0014136123936623335, + "learning_rate": 0.012201478480586513, + "loss": 0.0956, + "num_input_tokens_seen": 45147424, + "step": 34830 + }, + { + "epoch": 1.702074121124765, + "grad_norm": 0.001518469536677003, + "learning_rate": 0.012178218355587389, + "loss": 0.0698, + "num_input_tokens_seen": 45154432, + "step": 34835 + }, + { + "epoch": 1.7023184227884594, + "grad_norm": 0.001282695448026061, + "learning_rate": 0.01215497948443896, + "loss": 0.0598, + "num_input_tokens_seen": 45160800, + "step": 34840 + }, + { + "epoch": 1.7025627244521535, + "grad_norm": 0.000876336416695267, + "learning_rate": 0.012131761870724993, + "loss": 0.0814, + "num_input_tokens_seen": 45167744, + "step": 34845 + }, + { + "epoch": 1.7028070261158479, + "grad_norm": 0.0014577367110177875, + "learning_rate": 0.012108565518025893, + "loss": 0.0871, + "num_input_tokens_seen": 45174656, + "step": 34850 + }, + { + "epoch": 1.703051327779542, + "grad_norm": 0.0011425638804212213, + "learning_rate": 0.012085390429918862, + "loss": 0.0934, + "num_input_tokens_seen": 45181184, + "step": 34855 + }, + { + "epoch": 1.7032956294432364, + "grad_norm": 0.0017205625772476196, + "learning_rate": 0.012062236609977744, + "loss": 0.0837, + "num_input_tokens_seen": 45187936, + "step": 34860 + }, + { + "epoch": 1.7035399311069308, + "grad_norm": 0.0011964964214712381, + "learning_rate": 0.01203910406177318, + "loss": 0.0919, + "num_input_tokens_seen": 45194560, + "step": 34865 + }, + { + "epoch": 1.7037842327706252, + "grad_norm": 0.0009201011853292584, + "learning_rate": 0.01201599278887252, + "loss": 0.0601, + "num_input_tokens_seen": 45200800, + "step": 34870 + }, + { + "epoch": 1.7040285344343196, + "grad_norm": 0.0009048745268955827, + "learning_rate": 0.011992902794839744, + "loss": 0.0914, + "num_input_tokens_seen": 45207296, + "step": 34875 + }, + { + "epoch": 1.704272836098014, + "grad_norm": 0.0012901979498565197, + "learning_rate": 0.011969834083235703, + "loss": 0.0683, + "num_input_tokens_seen": 45213632, + "step": 34880 + }, + { + "epoch": 1.7045171377617083, + "grad_norm": 0.0026926547288894653, + "learning_rate": 0.011946786657617836, + "loss": 0.088, + "num_input_tokens_seen": 45220096, + "step": 34885 + }, + { + "epoch": 1.7047614394254025, + "grad_norm": 0.0017604152671992779, + "learning_rate": 0.011923760521540332, + "loss": 0.0699, + "num_input_tokens_seen": 45226560, + "step": 34890 + }, + { + "epoch": 1.7050057410890969, + "grad_norm": 0.0014257975853979588, + "learning_rate": 0.011900755678554153, + "loss": 0.0733, + "num_input_tokens_seen": 45232736, + "step": 34895 + }, + { + "epoch": 1.705250042752791, + "grad_norm": 0.0012757687363773584, + "learning_rate": 0.011877772132206893, + "loss": 0.0697, + "num_input_tokens_seen": 45239424, + "step": 34900 + }, + { + "epoch": 1.7054943444164854, + "grad_norm": 0.0013003140920773149, + "learning_rate": 0.011854809886042915, + "loss": 0.0769, + "num_input_tokens_seen": 45245504, + "step": 34905 + }, + { + "epoch": 1.7057386460801798, + "grad_norm": 0.0009462984744459391, + "learning_rate": 0.011831868943603325, + "loss": 0.063, + "num_input_tokens_seen": 45252256, + "step": 34910 + }, + { + "epoch": 1.7059829477438742, + "grad_norm": 0.001185135799460113, + "learning_rate": 0.011808949308425836, + "loss": 0.0912, + "num_input_tokens_seen": 45258208, + "step": 34915 + }, + { + "epoch": 1.7062272494075685, + "grad_norm": 0.0014795538736507297, + "learning_rate": 0.01178605098404501, + "loss": 0.1005, + "num_input_tokens_seen": 45264768, + "step": 34920 + }, + { + "epoch": 1.706471551071263, + "grad_norm": 0.001537772943265736, + "learning_rate": 0.011763173973992002, + "loss": 0.0608, + "num_input_tokens_seen": 45271520, + "step": 34925 + }, + { + "epoch": 1.706715852734957, + "grad_norm": 0.0008575518149882555, + "learning_rate": 0.011740318281794776, + "loss": 0.085, + "num_input_tokens_seen": 45277664, + "step": 34930 + }, + { + "epoch": 1.7069601543986515, + "grad_norm": 0.0017052239272743464, + "learning_rate": 0.01171748391097796, + "loss": 0.0864, + "num_input_tokens_seen": 45284192, + "step": 34935 + }, + { + "epoch": 1.7072044560623458, + "grad_norm": 0.0012216801987960935, + "learning_rate": 0.011694670865062873, + "loss": 0.1002, + "num_input_tokens_seen": 45290528, + "step": 34940 + }, + { + "epoch": 1.70744875772604, + "grad_norm": 0.001166549394838512, + "learning_rate": 0.011671879147567616, + "loss": 0.0918, + "num_input_tokens_seen": 45297024, + "step": 34945 + }, + { + "epoch": 1.7076930593897344, + "grad_norm": 0.001855909009464085, + "learning_rate": 0.011649108762006893, + "loss": 0.106, + "num_input_tokens_seen": 45303328, + "step": 34950 + }, + { + "epoch": 1.7079373610534287, + "grad_norm": 0.0010288192424923182, + "learning_rate": 0.011626359711892265, + "loss": 0.0876, + "num_input_tokens_seen": 45309696, + "step": 34955 + }, + { + "epoch": 1.7081816627171231, + "grad_norm": 0.00153722008690238, + "learning_rate": 0.01160363200073189, + "loss": 0.1046, + "num_input_tokens_seen": 45316224, + "step": 34960 + }, + { + "epoch": 1.7084259643808175, + "grad_norm": 0.0014934004284441471, + "learning_rate": 0.011580925632030614, + "loss": 0.1044, + "num_input_tokens_seen": 45322368, + "step": 34965 + }, + { + "epoch": 1.7086702660445119, + "grad_norm": 0.0010520222131162882, + "learning_rate": 0.011558240609290104, + "loss": 0.0782, + "num_input_tokens_seen": 45328704, + "step": 34970 + }, + { + "epoch": 1.708914567708206, + "grad_norm": 0.0016160978702828288, + "learning_rate": 0.011535576936008679, + "loss": 0.0935, + "num_input_tokens_seen": 45335328, + "step": 34975 + }, + { + "epoch": 1.7091588693719004, + "grad_norm": 0.0013447378296405077, + "learning_rate": 0.011512934615681309, + "loss": 0.088, + "num_input_tokens_seen": 45341696, + "step": 34980 + }, + { + "epoch": 1.7094031710355948, + "grad_norm": 0.0012808599276468158, + "learning_rate": 0.011490313651799765, + "loss": 0.0589, + "num_input_tokens_seen": 45348224, + "step": 34985 + }, + { + "epoch": 1.709647472699289, + "grad_norm": 0.0019280846463516355, + "learning_rate": 0.011467714047852512, + "loss": 0.0917, + "num_input_tokens_seen": 45354592, + "step": 34990 + }, + { + "epoch": 1.7098917743629833, + "grad_norm": 0.0014258805895224214, + "learning_rate": 0.011445135807324624, + "loss": 0.0669, + "num_input_tokens_seen": 45361152, + "step": 34995 + }, + { + "epoch": 1.7101360760266777, + "grad_norm": 0.0013722556177526712, + "learning_rate": 0.011422578933698002, + "loss": 0.076, + "num_input_tokens_seen": 45367936, + "step": 35000 + }, + { + "epoch": 1.7101360760266777, + "eval_loss": 0.08391347527503967, + "eval_runtime": 402.5655, + "eval_samples_per_second": 90.383, + "eval_steps_per_second": 22.598, + "num_input_tokens_seen": 45367936, + "step": 35000 + }, + { + "epoch": 1.710380377690372, + "grad_norm": 0.0018406867748126388, + "learning_rate": 0.011400043430451161, + "loss": 0.0833, + "num_input_tokens_seen": 45374048, + "step": 35005 + }, + { + "epoch": 1.7106246793540665, + "grad_norm": 0.0015916001284494996, + "learning_rate": 0.011377529301059392, + "loss": 0.0829, + "num_input_tokens_seen": 45380480, + "step": 35010 + }, + { + "epoch": 1.7108689810177609, + "grad_norm": 0.0022779095452278852, + "learning_rate": 0.011355036548994646, + "loss": 0.1072, + "num_input_tokens_seen": 45386912, + "step": 35015 + }, + { + "epoch": 1.711113282681455, + "grad_norm": 0.0012821053387597203, + "learning_rate": 0.011332565177725584, + "loss": 0.0687, + "num_input_tokens_seen": 45393344, + "step": 35020 + }, + { + "epoch": 1.7113575843451494, + "grad_norm": 0.0009763175039552152, + "learning_rate": 0.011310115190717585, + "loss": 0.0966, + "num_input_tokens_seen": 45399936, + "step": 35025 + }, + { + "epoch": 1.7116018860088436, + "grad_norm": 0.0008144594612531364, + "learning_rate": 0.01128768659143271, + "loss": 0.0638, + "num_input_tokens_seen": 45406368, + "step": 35030 + }, + { + "epoch": 1.711846187672538, + "grad_norm": 0.0009560482576489449, + "learning_rate": 0.011265279383329713, + "loss": 0.0754, + "num_input_tokens_seen": 45412608, + "step": 35035 + }, + { + "epoch": 1.7120904893362323, + "grad_norm": 0.0013823346234858036, + "learning_rate": 0.01124289356986411, + "loss": 0.0772, + "num_input_tokens_seen": 45419712, + "step": 35040 + }, + { + "epoch": 1.7123347909999267, + "grad_norm": 0.0012305593118071556, + "learning_rate": 0.011220529154488023, + "loss": 0.0884, + "num_input_tokens_seen": 45425952, + "step": 35045 + }, + { + "epoch": 1.712579092663621, + "grad_norm": 0.0012969761155545712, + "learning_rate": 0.011198186140650346, + "loss": 0.105, + "num_input_tokens_seen": 45432416, + "step": 35050 + }, + { + "epoch": 1.7128233943273155, + "grad_norm": 0.001119730295613408, + "learning_rate": 0.011175864531796685, + "loss": 0.0828, + "num_input_tokens_seen": 45439520, + "step": 35055 + }, + { + "epoch": 1.7130676959910098, + "grad_norm": 0.0020651621744036674, + "learning_rate": 0.011153564331369258, + "loss": 0.0845, + "num_input_tokens_seen": 45446112, + "step": 35060 + }, + { + "epoch": 1.713311997654704, + "grad_norm": 0.0011975087691098452, + "learning_rate": 0.011131285542807078, + "loss": 0.0822, + "num_input_tokens_seen": 45452736, + "step": 35065 + }, + { + "epoch": 1.7135562993183984, + "grad_norm": 0.0007124995463527739, + "learning_rate": 0.011109028169545815, + "loss": 0.0649, + "num_input_tokens_seen": 45459712, + "step": 35070 + }, + { + "epoch": 1.7138006009820925, + "grad_norm": 0.0020661239977926016, + "learning_rate": 0.011086792215017804, + "loss": 0.0779, + "num_input_tokens_seen": 45466496, + "step": 35075 + }, + { + "epoch": 1.714044902645787, + "grad_norm": 0.0018463021842762828, + "learning_rate": 0.011064577682652137, + "loss": 0.0793, + "num_input_tokens_seen": 45472768, + "step": 35080 + }, + { + "epoch": 1.7142892043094813, + "grad_norm": 0.00301506370306015, + "learning_rate": 0.011042384575874559, + "loss": 0.1045, + "num_input_tokens_seen": 45478912, + "step": 35085 + }, + { + "epoch": 1.7145335059731757, + "grad_norm": 0.0012282808311283588, + "learning_rate": 0.011020212898107512, + "loss": 0.0813, + "num_input_tokens_seen": 45485568, + "step": 35090 + }, + { + "epoch": 1.71477780763687, + "grad_norm": 0.0019491080893203616, + "learning_rate": 0.010998062652770197, + "loss": 0.0892, + "num_input_tokens_seen": 45491968, + "step": 35095 + }, + { + "epoch": 1.7150221093005644, + "grad_norm": 0.0012209564447402954, + "learning_rate": 0.010975933843278428, + "loss": 0.0713, + "num_input_tokens_seen": 45498240, + "step": 35100 + }, + { + "epoch": 1.7152664109642588, + "grad_norm": 0.0007464216905646026, + "learning_rate": 0.010953826473044714, + "loss": 0.0904, + "num_input_tokens_seen": 45504768, + "step": 35105 + }, + { + "epoch": 1.715510712627953, + "grad_norm": 0.0007407630328088999, + "learning_rate": 0.010931740545478357, + "loss": 0.0776, + "num_input_tokens_seen": 45511008, + "step": 35110 + }, + { + "epoch": 1.7157550142916473, + "grad_norm": 0.001019757124595344, + "learning_rate": 0.010909676063985218, + "loss": 0.064, + "num_input_tokens_seen": 45517696, + "step": 35115 + }, + { + "epoch": 1.7159993159553415, + "grad_norm": 0.0010635664220899343, + "learning_rate": 0.010887633031967974, + "loss": 0.0532, + "num_input_tokens_seen": 45524224, + "step": 35120 + }, + { + "epoch": 1.7162436176190359, + "grad_norm": 0.0016556333284825087, + "learning_rate": 0.01086561145282589, + "loss": 0.0886, + "num_input_tokens_seen": 45530944, + "step": 35125 + }, + { + "epoch": 1.7164879192827303, + "grad_norm": 0.0011274394346401095, + "learning_rate": 0.010843611329954983, + "loss": 0.0675, + "num_input_tokens_seen": 45537600, + "step": 35130 + }, + { + "epoch": 1.7167322209464246, + "grad_norm": 0.0013892611023038626, + "learning_rate": 0.010821632666747988, + "loss": 0.1102, + "num_input_tokens_seen": 45544064, + "step": 35135 + }, + { + "epoch": 1.716976522610119, + "grad_norm": 0.001074198866263032, + "learning_rate": 0.010799675466594244, + "loss": 0.0697, + "num_input_tokens_seen": 45550624, + "step": 35140 + }, + { + "epoch": 1.7172208242738134, + "grad_norm": 0.0020507851149886847, + "learning_rate": 0.010777739732879826, + "loss": 0.0566, + "num_input_tokens_seen": 45557280, + "step": 35145 + }, + { + "epoch": 1.7174651259375078, + "grad_norm": 0.0015218022745102644, + "learning_rate": 0.010755825468987562, + "loss": 0.0752, + "num_input_tokens_seen": 45563136, + "step": 35150 + }, + { + "epoch": 1.717709427601202, + "grad_norm": 0.0016525281826034188, + "learning_rate": 0.010733932678296814, + "loss": 0.081, + "num_input_tokens_seen": 45569440, + "step": 35155 + }, + { + "epoch": 1.7179537292648963, + "grad_norm": 0.00146756402682513, + "learning_rate": 0.010712061364183817, + "loss": 0.0912, + "num_input_tokens_seen": 45575552, + "step": 35160 + }, + { + "epoch": 1.7181980309285905, + "grad_norm": 0.0013646676670759916, + "learning_rate": 0.010690211530021337, + "loss": 0.1031, + "num_input_tokens_seen": 45581760, + "step": 35165 + }, + { + "epoch": 1.7184423325922848, + "grad_norm": 0.0014010875020176172, + "learning_rate": 0.01066838317917893, + "loss": 0.0939, + "num_input_tokens_seen": 45588128, + "step": 35170 + }, + { + "epoch": 1.7186866342559792, + "grad_norm": 0.0015158734750002623, + "learning_rate": 0.010646576315022787, + "loss": 0.0806, + "num_input_tokens_seen": 45594752, + "step": 35175 + }, + { + "epoch": 1.7189309359196736, + "grad_norm": 0.001699213171377778, + "learning_rate": 0.010624790940915785, + "loss": 0.099, + "num_input_tokens_seen": 45601152, + "step": 35180 + }, + { + "epoch": 1.719175237583368, + "grad_norm": 0.0014908502344042063, + "learning_rate": 0.0106030270602175, + "loss": 0.1023, + "num_input_tokens_seen": 45607616, + "step": 35185 + }, + { + "epoch": 1.7194195392470624, + "grad_norm": 0.0005800547078251839, + "learning_rate": 0.010581284676284252, + "loss": 0.0763, + "num_input_tokens_seen": 45614240, + "step": 35190 + }, + { + "epoch": 1.7196638409107567, + "grad_norm": 0.0011018238728865981, + "learning_rate": 0.010559563792468923, + "loss": 0.063, + "num_input_tokens_seen": 45620576, + "step": 35195 + }, + { + "epoch": 1.719908142574451, + "grad_norm": 0.0008528809994459152, + "learning_rate": 0.010537864412121217, + "loss": 0.0653, + "num_input_tokens_seen": 45627104, + "step": 35200 + }, + { + "epoch": 1.719908142574451, + "eval_loss": 0.08412104845046997, + "eval_runtime": 402.3055, + "eval_samples_per_second": 90.441, + "eval_steps_per_second": 22.612, + "num_input_tokens_seen": 45627104, + "step": 35200 + }, + { + "epoch": 1.7201524442381453, + "grad_norm": 0.0014780090423300862, + "learning_rate": 0.010516186538587357, + "loss": 0.0916, + "num_input_tokens_seen": 45633408, + "step": 35205 + }, + { + "epoch": 1.7203967459018394, + "grad_norm": 0.0009122187620960176, + "learning_rate": 0.01049453017521042, + "loss": 0.0656, + "num_input_tokens_seen": 45639552, + "step": 35210 + }, + { + "epoch": 1.7206410475655338, + "grad_norm": 0.00289192795753479, + "learning_rate": 0.010472895325330083, + "loss": 0.092, + "num_input_tokens_seen": 45646432, + "step": 35215 + }, + { + "epoch": 1.7208853492292282, + "grad_norm": 0.0016199011588469148, + "learning_rate": 0.010451281992282662, + "loss": 0.0912, + "num_input_tokens_seen": 45652928, + "step": 35220 + }, + { + "epoch": 1.7211296508929226, + "grad_norm": 0.002079498255625367, + "learning_rate": 0.01042969017940124, + "loss": 0.0802, + "num_input_tokens_seen": 45659072, + "step": 35225 + }, + { + "epoch": 1.721373952556617, + "grad_norm": 0.0023277837317436934, + "learning_rate": 0.01040811989001557, + "loss": 0.0944, + "num_input_tokens_seen": 45665408, + "step": 35230 + }, + { + "epoch": 1.7216182542203113, + "grad_norm": 0.0010268694022670388, + "learning_rate": 0.010386571127451992, + "loss": 0.0581, + "num_input_tokens_seen": 45671584, + "step": 35235 + }, + { + "epoch": 1.7218625558840057, + "grad_norm": 0.0011801713844761252, + "learning_rate": 0.010365043895033682, + "loss": 0.0736, + "num_input_tokens_seen": 45677920, + "step": 35240 + }, + { + "epoch": 1.7221068575476999, + "grad_norm": 0.0015021548606455326, + "learning_rate": 0.010343538196080365, + "loss": 0.0773, + "num_input_tokens_seen": 45684512, + "step": 35245 + }, + { + "epoch": 1.7223511592113943, + "grad_norm": 0.001615706947632134, + "learning_rate": 0.010322054033908457, + "loss": 0.0611, + "num_input_tokens_seen": 45690816, + "step": 35250 + }, + { + "epoch": 1.7225954608750884, + "grad_norm": 0.002637833822518587, + "learning_rate": 0.010300591411831156, + "loss": 0.1168, + "num_input_tokens_seen": 45697504, + "step": 35255 + }, + { + "epoch": 1.7228397625387828, + "grad_norm": 0.001132380566559732, + "learning_rate": 0.010279150333158198, + "loss": 0.092, + "num_input_tokens_seen": 45703744, + "step": 35260 + }, + { + "epoch": 1.7230840642024772, + "grad_norm": 0.0028297072276473045, + "learning_rate": 0.010257730801196107, + "loss": 0.0836, + "num_input_tokens_seen": 45711008, + "step": 35265 + }, + { + "epoch": 1.7233283658661716, + "grad_norm": 0.0013911889400333166, + "learning_rate": 0.010236332819248056, + "loss": 0.0655, + "num_input_tokens_seen": 45717760, + "step": 35270 + }, + { + "epoch": 1.723572667529866, + "grad_norm": 0.0011898305965587497, + "learning_rate": 0.010214956390613854, + "loss": 0.0863, + "num_input_tokens_seen": 45724032, + "step": 35275 + }, + { + "epoch": 1.7238169691935603, + "grad_norm": 0.0012647039256989956, + "learning_rate": 0.010193601518590034, + "loss": 0.0743, + "num_input_tokens_seen": 45730144, + "step": 35280 + }, + { + "epoch": 1.7240612708572547, + "grad_norm": 0.0012395471567288041, + "learning_rate": 0.010172268206469758, + "loss": 0.0872, + "num_input_tokens_seen": 45736128, + "step": 35285 + }, + { + "epoch": 1.7243055725209488, + "grad_norm": 0.0011586908949539065, + "learning_rate": 0.010150956457542897, + "loss": 0.0833, + "num_input_tokens_seen": 45742368, + "step": 35290 + }, + { + "epoch": 1.7245498741846432, + "grad_norm": 0.0011439700610935688, + "learning_rate": 0.010129666275096054, + "loss": 0.0824, + "num_input_tokens_seen": 45748896, + "step": 35295 + }, + { + "epoch": 1.7247941758483374, + "grad_norm": 0.000804560782853514, + "learning_rate": 0.010108397662412338, + "loss": 0.0686, + "num_input_tokens_seen": 45755360, + "step": 35300 + }, + { + "epoch": 1.7250384775120318, + "grad_norm": 0.001886901562102139, + "learning_rate": 0.010087150622771707, + "loss": 0.0963, + "num_input_tokens_seen": 45762112, + "step": 35305 + }, + { + "epoch": 1.7252827791757261, + "grad_norm": 0.0008513766806572676, + "learning_rate": 0.010065925159450739, + "loss": 0.0571, + "num_input_tokens_seen": 45768480, + "step": 35310 + }, + { + "epoch": 1.7255270808394205, + "grad_norm": 0.0007233843789435923, + "learning_rate": 0.010044721275722618, + "loss": 0.0581, + "num_input_tokens_seen": 45775264, + "step": 35315 + }, + { + "epoch": 1.725771382503115, + "grad_norm": 0.0005638026050291955, + "learning_rate": 0.01002353897485726, + "loss": 0.0815, + "num_input_tokens_seen": 45781760, + "step": 35320 + }, + { + "epoch": 1.7260156841668093, + "grad_norm": 0.0006096758297644556, + "learning_rate": 0.010002378260121236, + "loss": 0.0591, + "num_input_tokens_seen": 45788256, + "step": 35325 + }, + { + "epoch": 1.7262599858305037, + "grad_norm": 0.0016877492889761925, + "learning_rate": 0.009981239134777786, + "loss": 0.0904, + "num_input_tokens_seen": 45794816, + "step": 35330 + }, + { + "epoch": 1.7265042874941978, + "grad_norm": 0.002083060098811984, + "learning_rate": 0.009960121602086884, + "loss": 0.0828, + "num_input_tokens_seen": 45801088, + "step": 35335 + }, + { + "epoch": 1.7267485891578922, + "grad_norm": 0.0010364012559875846, + "learning_rate": 0.009939025665305062, + "loss": 0.0994, + "num_input_tokens_seen": 45807648, + "step": 35340 + }, + { + "epoch": 1.7269928908215864, + "grad_norm": 0.0013264797162264585, + "learning_rate": 0.009917951327685597, + "loss": 0.0909, + "num_input_tokens_seen": 45813760, + "step": 35345 + }, + { + "epoch": 1.7272371924852807, + "grad_norm": 0.0012034856481477618, + "learning_rate": 0.009896898592478425, + "loss": 0.0526, + "num_input_tokens_seen": 45820608, + "step": 35350 + }, + { + "epoch": 1.7274814941489751, + "grad_norm": 0.0009372577187605202, + "learning_rate": 0.009875867462930132, + "loss": 0.081, + "num_input_tokens_seen": 45826816, + "step": 35355 + }, + { + "epoch": 1.7277257958126695, + "grad_norm": 0.0019965232349932194, + "learning_rate": 0.009854857942284006, + "loss": 0.0973, + "num_input_tokens_seen": 45833216, + "step": 35360 + }, + { + "epoch": 1.7279700974763639, + "grad_norm": 0.001757569843903184, + "learning_rate": 0.009833870033779923, + "loss": 0.0914, + "num_input_tokens_seen": 45839712, + "step": 35365 + }, + { + "epoch": 1.7282143991400583, + "grad_norm": 0.0013563822722062469, + "learning_rate": 0.009812903740654527, + "loss": 0.0871, + "num_input_tokens_seen": 45846208, + "step": 35370 + }, + { + "epoch": 1.7284587008037526, + "grad_norm": 0.0014223260805010796, + "learning_rate": 0.009791959066141097, + "loss": 0.1138, + "num_input_tokens_seen": 45852512, + "step": 35375 + }, + { + "epoch": 1.7287030024674468, + "grad_norm": 0.0018812059424817562, + "learning_rate": 0.009771036013469537, + "loss": 0.0779, + "num_input_tokens_seen": 45859104, + "step": 35380 + }, + { + "epoch": 1.7289473041311412, + "grad_norm": 0.0013910079142078757, + "learning_rate": 0.00975013458586646, + "loss": 0.0794, + "num_input_tokens_seen": 45865568, + "step": 35385 + }, + { + "epoch": 1.7291916057948353, + "grad_norm": 0.0011515849037095904, + "learning_rate": 0.009729254786555107, + "loss": 0.0824, + "num_input_tokens_seen": 45872224, + "step": 35390 + }, + { + "epoch": 1.7294359074585297, + "grad_norm": 0.002680201083421707, + "learning_rate": 0.009708396618755421, + "loss": 0.1064, + "num_input_tokens_seen": 45878752, + "step": 35395 + }, + { + "epoch": 1.729680209122224, + "grad_norm": 0.001171499490737915, + "learning_rate": 0.009687560085683994, + "loss": 0.0772, + "num_input_tokens_seen": 45885312, + "step": 35400 + }, + { + "epoch": 1.729680209122224, + "eval_loss": 0.08384579420089722, + "eval_runtime": 401.9629, + "eval_samples_per_second": 90.518, + "eval_steps_per_second": 22.631, + "num_input_tokens_seen": 45885312, + "step": 35400 + }, + { + "epoch": 1.7299245107859185, + "grad_norm": 0.0015026816399767995, + "learning_rate": 0.009666745190554054, + "loss": 0.0585, + "num_input_tokens_seen": 45891872, + "step": 35405 + }, + { + "epoch": 1.7301688124496128, + "grad_norm": 0.001532172434963286, + "learning_rate": 0.009645951936575553, + "loss": 0.0844, + "num_input_tokens_seen": 45898240, + "step": 35410 + }, + { + "epoch": 1.7304131141133072, + "grad_norm": 0.0014860235387459397, + "learning_rate": 0.00962518032695509, + "loss": 0.0842, + "num_input_tokens_seen": 45904544, + "step": 35415 + }, + { + "epoch": 1.7306574157770016, + "grad_norm": 0.0010537131456658244, + "learning_rate": 0.009604430364895855, + "loss": 0.0783, + "num_input_tokens_seen": 45910848, + "step": 35420 + }, + { + "epoch": 1.7309017174406958, + "grad_norm": 0.0012164368527010083, + "learning_rate": 0.00958370205359777, + "loss": 0.0636, + "num_input_tokens_seen": 45917600, + "step": 35425 + }, + { + "epoch": 1.7311460191043901, + "grad_norm": 0.001758026541210711, + "learning_rate": 0.009562995396257445, + "loss": 0.0837, + "num_input_tokens_seen": 45924160, + "step": 35430 + }, + { + "epoch": 1.7313903207680843, + "grad_norm": 0.0008391918381676078, + "learning_rate": 0.009542310396068026, + "loss": 0.0615, + "num_input_tokens_seen": 45930880, + "step": 35435 + }, + { + "epoch": 1.7316346224317787, + "grad_norm": 0.002937317593023181, + "learning_rate": 0.009521647056219495, + "loss": 0.0778, + "num_input_tokens_seen": 45937280, + "step": 35440 + }, + { + "epoch": 1.731878924095473, + "grad_norm": 0.002063825959339738, + "learning_rate": 0.00950100537989832, + "loss": 0.0713, + "num_input_tokens_seen": 45943872, + "step": 35445 + }, + { + "epoch": 1.7321232257591674, + "grad_norm": 0.001692155608907342, + "learning_rate": 0.00948038537028772, + "loss": 0.0769, + "num_input_tokens_seen": 45950112, + "step": 35450 + }, + { + "epoch": 1.7323675274228618, + "grad_norm": 0.0009563821367919445, + "learning_rate": 0.009459787030567617, + "loss": 0.0656, + "num_input_tokens_seen": 45956896, + "step": 35455 + }, + { + "epoch": 1.7326118290865562, + "grad_norm": 0.0014286597725003958, + "learning_rate": 0.00943921036391449, + "loss": 0.0763, + "num_input_tokens_seen": 45963296, + "step": 35460 + }, + { + "epoch": 1.7328561307502504, + "grad_norm": 0.0018598936730995774, + "learning_rate": 0.009418655373501483, + "loss": 0.1077, + "num_input_tokens_seen": 45970016, + "step": 35465 + }, + { + "epoch": 1.7331004324139447, + "grad_norm": 0.0017558166291564703, + "learning_rate": 0.00939812206249851, + "loss": 0.0863, + "num_input_tokens_seen": 45976416, + "step": 35470 + }, + { + "epoch": 1.7333447340776391, + "grad_norm": 0.0010708929039537907, + "learning_rate": 0.009377610434072004, + "loss": 0.067, + "num_input_tokens_seen": 45982848, + "step": 35475 + }, + { + "epoch": 1.7335890357413333, + "grad_norm": 0.0010590238962322474, + "learning_rate": 0.009357120491385167, + "loss": 0.065, + "num_input_tokens_seen": 45989536, + "step": 35480 + }, + { + "epoch": 1.7338333374050277, + "grad_norm": 0.0022553761955350637, + "learning_rate": 0.009336652237597743, + "loss": 0.1066, + "num_input_tokens_seen": 45995968, + "step": 35485 + }, + { + "epoch": 1.734077639068722, + "grad_norm": 0.0022920058108866215, + "learning_rate": 0.009316205675866251, + "loss": 0.0578, + "num_input_tokens_seen": 46002208, + "step": 35490 + }, + { + "epoch": 1.7343219407324164, + "grad_norm": 0.001498801400884986, + "learning_rate": 0.00929578080934379, + "loss": 0.0746, + "num_input_tokens_seen": 46008896, + "step": 35495 + }, + { + "epoch": 1.7345662423961108, + "grad_norm": 0.0016018160386011004, + "learning_rate": 0.00927537764118012, + "loss": 0.0823, + "num_input_tokens_seen": 46015392, + "step": 35500 + }, + { + "epoch": 1.7348105440598052, + "grad_norm": 0.0008559554116800427, + "learning_rate": 0.009254996174521678, + "loss": 0.0731, + "num_input_tokens_seen": 46021856, + "step": 35505 + }, + { + "epoch": 1.7350548457234993, + "grad_norm": 0.001222031656652689, + "learning_rate": 0.009234636412511531, + "loss": 0.0763, + "num_input_tokens_seen": 46028416, + "step": 35510 + }, + { + "epoch": 1.7352991473871937, + "grad_norm": 0.000876223377417773, + "learning_rate": 0.009214298358289418, + "loss": 0.0715, + "num_input_tokens_seen": 46035072, + "step": 35515 + }, + { + "epoch": 1.735543449050888, + "grad_norm": 0.0017351371934637427, + "learning_rate": 0.00919398201499173, + "loss": 0.0798, + "num_input_tokens_seen": 46041344, + "step": 35520 + }, + { + "epoch": 1.7357877507145822, + "grad_norm": 0.0010547611163929105, + "learning_rate": 0.009173687385751495, + "loss": 0.0856, + "num_input_tokens_seen": 46047552, + "step": 35525 + }, + { + "epoch": 1.7360320523782766, + "grad_norm": 0.0010680605191737413, + "learning_rate": 0.009153414473698407, + "loss": 0.0863, + "num_input_tokens_seen": 46054432, + "step": 35530 + }, + { + "epoch": 1.736276354041971, + "grad_norm": 0.0015947475330904126, + "learning_rate": 0.009133163281958784, + "loss": 0.0872, + "num_input_tokens_seen": 46061056, + "step": 35535 + }, + { + "epoch": 1.7365206557056654, + "grad_norm": 0.0015367965679615736, + "learning_rate": 0.009112933813655627, + "loss": 0.0708, + "num_input_tokens_seen": 46067776, + "step": 35540 + }, + { + "epoch": 1.7367649573693598, + "grad_norm": 0.0018777980003505945, + "learning_rate": 0.009092726071908573, + "loss": 0.0865, + "num_input_tokens_seen": 46074144, + "step": 35545 + }, + { + "epoch": 1.7370092590330541, + "grad_norm": 0.002015543868765235, + "learning_rate": 0.0090725400598339, + "loss": 0.0698, + "num_input_tokens_seen": 46080480, + "step": 35550 + }, + { + "epoch": 1.7372535606967483, + "grad_norm": 0.0014131757197901607, + "learning_rate": 0.009052375780544563, + "loss": 0.0726, + "num_input_tokens_seen": 46086880, + "step": 35555 + }, + { + "epoch": 1.7374978623604427, + "grad_norm": 0.001138562336564064, + "learning_rate": 0.009032233237150144, + "loss": 0.0763, + "num_input_tokens_seen": 46093408, + "step": 35560 + }, + { + "epoch": 1.737742164024137, + "grad_norm": 0.0011837418423965573, + "learning_rate": 0.009012112432756875, + "loss": 0.1006, + "num_input_tokens_seen": 46099776, + "step": 35565 + }, + { + "epoch": 1.7379864656878312, + "grad_norm": 0.001744271838106215, + "learning_rate": 0.008992013370467605, + "loss": 0.0803, + "num_input_tokens_seen": 46106240, + "step": 35570 + }, + { + "epoch": 1.7382307673515256, + "grad_norm": 0.0009435852989554405, + "learning_rate": 0.008971936053381924, + "loss": 0.0627, + "num_input_tokens_seen": 46112736, + "step": 35575 + }, + { + "epoch": 1.73847506901522, + "grad_norm": 0.001070657977834344, + "learning_rate": 0.008951880484595953, + "loss": 0.0672, + "num_input_tokens_seen": 46119200, + "step": 35580 + }, + { + "epoch": 1.7387193706789144, + "grad_norm": 0.001443561282940209, + "learning_rate": 0.008931846667202552, + "loss": 0.0539, + "num_input_tokens_seen": 46125824, + "step": 35585 + }, + { + "epoch": 1.7389636723426087, + "grad_norm": 0.0011702593183144927, + "learning_rate": 0.008911834604291152, + "loss": 0.0862, + "num_input_tokens_seen": 46132384, + "step": 35590 + }, + { + "epoch": 1.7392079740063031, + "grad_norm": 0.0014862708048895001, + "learning_rate": 0.008891844298947882, + "loss": 0.0773, + "num_input_tokens_seen": 46138880, + "step": 35595 + }, + { + "epoch": 1.7394522756699973, + "grad_norm": 0.0017970436019822955, + "learning_rate": 0.008871875754255508, + "loss": 0.069, + "num_input_tokens_seen": 46145568, + "step": 35600 + }, + { + "epoch": 1.7394522756699973, + "eval_loss": 0.0841720774769783, + "eval_runtime": 402.4607, + "eval_samples_per_second": 90.406, + "eval_steps_per_second": 22.603, + "num_input_tokens_seen": 46145568, + "step": 35600 + }, + { + "epoch": 1.7396965773336917, + "grad_norm": 0.00217668735422194, + "learning_rate": 0.008851928973293422, + "loss": 0.071, + "num_input_tokens_seen": 46152448, + "step": 35605 + }, + { + "epoch": 1.7399408789973858, + "grad_norm": 0.0010959503706544638, + "learning_rate": 0.00883200395913764, + "loss": 0.092, + "num_input_tokens_seen": 46158880, + "step": 35610 + }, + { + "epoch": 1.7401851806610802, + "grad_norm": 0.0009396210662089288, + "learning_rate": 0.00881210071486091, + "loss": 0.0726, + "num_input_tokens_seen": 46165664, + "step": 35615 + }, + { + "epoch": 1.7404294823247746, + "grad_norm": 0.0010263427393510938, + "learning_rate": 0.008792219243532505, + "loss": 0.0786, + "num_input_tokens_seen": 46172384, + "step": 35620 + }, + { + "epoch": 1.740673783988469, + "grad_norm": 0.0022283531725406647, + "learning_rate": 0.008772359548218428, + "loss": 0.0849, + "num_input_tokens_seen": 46179104, + "step": 35625 + }, + { + "epoch": 1.7409180856521633, + "grad_norm": 0.0015292185125872493, + "learning_rate": 0.008752521631981274, + "loss": 0.0608, + "num_input_tokens_seen": 46185728, + "step": 35630 + }, + { + "epoch": 1.7411623873158577, + "grad_norm": 0.00122721993830055, + "learning_rate": 0.008732705497880315, + "loss": 0.0878, + "num_input_tokens_seen": 46192192, + "step": 35635 + }, + { + "epoch": 1.741406688979552, + "grad_norm": 0.0010112345917150378, + "learning_rate": 0.008712911148971459, + "loss": 0.0528, + "num_input_tokens_seen": 46198656, + "step": 35640 + }, + { + "epoch": 1.7416509906432462, + "grad_norm": 0.0017946050502359867, + "learning_rate": 0.008693138588307208, + "loss": 0.0874, + "num_input_tokens_seen": 46204928, + "step": 35645 + }, + { + "epoch": 1.7418952923069406, + "grad_norm": 0.0009994895663112402, + "learning_rate": 0.008673387818936762, + "loss": 0.0781, + "num_input_tokens_seen": 46211136, + "step": 35650 + }, + { + "epoch": 1.7421395939706348, + "grad_norm": 0.001482008839957416, + "learning_rate": 0.008653658843905948, + "loss": 0.0807, + "num_input_tokens_seen": 46217728, + "step": 35655 + }, + { + "epoch": 1.7423838956343292, + "grad_norm": 0.0013500809436663985, + "learning_rate": 0.0086339516662572, + "loss": 0.1025, + "num_input_tokens_seen": 46224256, + "step": 35660 + }, + { + "epoch": 1.7426281972980235, + "grad_norm": 0.0012570901308208704, + "learning_rate": 0.008614266289029638, + "loss": 0.0655, + "num_input_tokens_seen": 46230816, + "step": 35665 + }, + { + "epoch": 1.742872498961718, + "grad_norm": 0.0016506962710991502, + "learning_rate": 0.008594602715258965, + "loss": 0.0837, + "num_input_tokens_seen": 46237248, + "step": 35670 + }, + { + "epoch": 1.7431168006254123, + "grad_norm": 0.0010252740466967225, + "learning_rate": 0.008574960947977573, + "loss": 0.078, + "num_input_tokens_seen": 46244352, + "step": 35675 + }, + { + "epoch": 1.7433611022891067, + "grad_norm": 0.0012585383374243975, + "learning_rate": 0.008555340990214438, + "loss": 0.0789, + "num_input_tokens_seen": 46251040, + "step": 35680 + }, + { + "epoch": 1.743605403952801, + "grad_norm": 0.001618091482669115, + "learning_rate": 0.008535742844995258, + "loss": 0.0715, + "num_input_tokens_seen": 46257920, + "step": 35685 + }, + { + "epoch": 1.7438497056164952, + "grad_norm": 0.0015631035203114152, + "learning_rate": 0.008516166515342266, + "loss": 0.0936, + "num_input_tokens_seen": 46264512, + "step": 35690 + }, + { + "epoch": 1.7440940072801896, + "grad_norm": 0.001904514734633267, + "learning_rate": 0.008496612004274411, + "loss": 0.0702, + "num_input_tokens_seen": 46270976, + "step": 35695 + }, + { + "epoch": 1.7443383089438838, + "grad_norm": 0.002117785392329097, + "learning_rate": 0.008477079314807201, + "loss": 0.0816, + "num_input_tokens_seen": 46277696, + "step": 35700 + }, + { + "epoch": 1.7445826106075781, + "grad_norm": 0.0008722046623006463, + "learning_rate": 0.008457568449952874, + "loss": 0.0572, + "num_input_tokens_seen": 46283968, + "step": 35705 + }, + { + "epoch": 1.7448269122712725, + "grad_norm": 0.001417159684933722, + "learning_rate": 0.008438079412720189, + "loss": 0.0752, + "num_input_tokens_seen": 46290624, + "step": 35710 + }, + { + "epoch": 1.745071213934967, + "grad_norm": 0.0014709942042827606, + "learning_rate": 0.00841861220611466, + "loss": 0.0724, + "num_input_tokens_seen": 46297472, + "step": 35715 + }, + { + "epoch": 1.7453155155986613, + "grad_norm": 0.001670663245022297, + "learning_rate": 0.008399166833138355, + "loss": 0.0876, + "num_input_tokens_seen": 46303936, + "step": 35720 + }, + { + "epoch": 1.7455598172623557, + "grad_norm": 0.001772728399373591, + "learning_rate": 0.008379743296789987, + "loss": 0.0792, + "num_input_tokens_seen": 46310144, + "step": 35725 + }, + { + "epoch": 1.74580411892605, + "grad_norm": 0.0012480865698307753, + "learning_rate": 0.008360341600064896, + "loss": 0.094, + "num_input_tokens_seen": 46317216, + "step": 35730 + }, + { + "epoch": 1.7460484205897442, + "grad_norm": 0.0013591385213658214, + "learning_rate": 0.008340961745955121, + "loss": 0.0652, + "num_input_tokens_seen": 46323712, + "step": 35735 + }, + { + "epoch": 1.7462927222534386, + "grad_norm": 0.001528942841105163, + "learning_rate": 0.008321603737449224, + "loss": 0.0636, + "num_input_tokens_seen": 46330592, + "step": 35740 + }, + { + "epoch": 1.7465370239171327, + "grad_norm": 0.0015212433645501733, + "learning_rate": 0.008302267577532479, + "loss": 0.0751, + "num_input_tokens_seen": 46336832, + "step": 35745 + }, + { + "epoch": 1.746781325580827, + "grad_norm": 0.0014953040517866611, + "learning_rate": 0.008282953269186771, + "loss": 0.0699, + "num_input_tokens_seen": 46343424, + "step": 35750 + }, + { + "epoch": 1.7470256272445215, + "grad_norm": 0.0013409927487373352, + "learning_rate": 0.008263660815390567, + "loss": 0.0684, + "num_input_tokens_seen": 46349760, + "step": 35755 + }, + { + "epoch": 1.7472699289082159, + "grad_norm": 0.0009070130181498826, + "learning_rate": 0.008244390219119069, + "loss": 0.0807, + "num_input_tokens_seen": 46356704, + "step": 35760 + }, + { + "epoch": 1.7475142305719102, + "grad_norm": 0.0017160861752927303, + "learning_rate": 0.008225141483343967, + "loss": 0.0942, + "num_input_tokens_seen": 46362912, + "step": 35765 + }, + { + "epoch": 1.7477585322356046, + "grad_norm": 0.001523882383480668, + "learning_rate": 0.00820591461103372, + "loss": 0.0819, + "num_input_tokens_seen": 46369376, + "step": 35770 + }, + { + "epoch": 1.748002833899299, + "grad_norm": 0.0012044035829603672, + "learning_rate": 0.008186709605153358, + "loss": 0.0772, + "num_input_tokens_seen": 46375872, + "step": 35775 + }, + { + "epoch": 1.7482471355629932, + "grad_norm": 0.0034429761581122875, + "learning_rate": 0.008167526468664492, + "loss": 0.1172, + "num_input_tokens_seen": 46382688, + "step": 35780 + }, + { + "epoch": 1.7484914372266875, + "grad_norm": 0.0011300368933007121, + "learning_rate": 0.008148365204525443, + "loss": 0.1024, + "num_input_tokens_seen": 46389344, + "step": 35785 + }, + { + "epoch": 1.7487357388903817, + "grad_norm": 0.0005213282420299947, + "learning_rate": 0.00812922581569106, + "loss": 0.0599, + "num_input_tokens_seen": 46396480, + "step": 35790 + }, + { + "epoch": 1.748980040554076, + "grad_norm": 0.0014429041184484959, + "learning_rate": 0.008110108305112934, + "loss": 0.0822, + "num_input_tokens_seen": 46403040, + "step": 35795 + }, + { + "epoch": 1.7492243422177705, + "grad_norm": 0.0013050915440544486, + "learning_rate": 0.008091012675739223, + "loss": 0.0938, + "num_input_tokens_seen": 46409504, + "step": 35800 + }, + { + "epoch": 1.7492243422177705, + "eval_loss": 0.08384208381175995, + "eval_runtime": 402.1506, + "eval_samples_per_second": 90.476, + "eval_steps_per_second": 22.621, + "num_input_tokens_seen": 46409504, + "step": 35800 + }, + { + "epoch": 1.7494686438814648, + "grad_norm": 0.0015843694563955069, + "learning_rate": 0.008071938930514671, + "loss": 0.0891, + "num_input_tokens_seen": 46416160, + "step": 35805 + }, + { + "epoch": 1.7497129455451592, + "grad_norm": 0.0010498407064005733, + "learning_rate": 0.008052887072380726, + "loss": 0.0548, + "num_input_tokens_seen": 46422656, + "step": 35810 + }, + { + "epoch": 1.7499572472088536, + "grad_norm": 0.0009820351842790842, + "learning_rate": 0.008033857104275437, + "loss": 0.0694, + "num_input_tokens_seen": 46429152, + "step": 35815 + }, + { + "epoch": 1.750201548872548, + "grad_norm": 0.0015653979498893023, + "learning_rate": 0.008014849029133424, + "loss": 0.0899, + "num_input_tokens_seen": 46435136, + "step": 35820 + }, + { + "epoch": 1.7504458505362421, + "grad_norm": 0.0010624080896377563, + "learning_rate": 0.007995862849885975, + "loss": 0.0853, + "num_input_tokens_seen": 46441760, + "step": 35825 + }, + { + "epoch": 1.7506901521999365, + "grad_norm": 0.0017882823012769222, + "learning_rate": 0.007976898569461032, + "loss": 0.0805, + "num_input_tokens_seen": 46447936, + "step": 35830 + }, + { + "epoch": 1.7509344538636307, + "grad_norm": 0.0013126898556947708, + "learning_rate": 0.007957956190783088, + "loss": 0.0696, + "num_input_tokens_seen": 46454304, + "step": 35835 + }, + { + "epoch": 1.751178755527325, + "grad_norm": 0.0015986233483999968, + "learning_rate": 0.007939035716773324, + "loss": 0.0955, + "num_input_tokens_seen": 46460832, + "step": 35840 + }, + { + "epoch": 1.7514230571910194, + "grad_norm": 0.0012254557805135846, + "learning_rate": 0.007920137150349487, + "loss": 0.0761, + "num_input_tokens_seen": 46467200, + "step": 35845 + }, + { + "epoch": 1.7516673588547138, + "grad_norm": 0.001495607546530664, + "learning_rate": 0.007901260494425981, + "loss": 0.0785, + "num_input_tokens_seen": 46473824, + "step": 35850 + }, + { + "epoch": 1.7519116605184082, + "grad_norm": 0.0014342815848067403, + "learning_rate": 0.007882405751913861, + "loss": 0.0935, + "num_input_tokens_seen": 46480352, + "step": 35855 + }, + { + "epoch": 1.7521559621821026, + "grad_norm": 0.000883500964846462, + "learning_rate": 0.007863572925720702, + "loss": 0.0792, + "num_input_tokens_seen": 46486656, + "step": 35860 + }, + { + "epoch": 1.752400263845797, + "grad_norm": 0.0015212730504572392, + "learning_rate": 0.007844762018750827, + "loss": 0.1244, + "num_input_tokens_seen": 46492736, + "step": 35865 + }, + { + "epoch": 1.752644565509491, + "grad_norm": 0.0007547753630205989, + "learning_rate": 0.007825973033905054, + "loss": 0.0778, + "num_input_tokens_seen": 46499168, + "step": 35870 + }, + { + "epoch": 1.7528888671731855, + "grad_norm": 0.0012346981093287468, + "learning_rate": 0.007807205974080927, + "loss": 0.0781, + "num_input_tokens_seen": 46505952, + "step": 35875 + }, + { + "epoch": 1.7531331688368796, + "grad_norm": 0.001034785294905305, + "learning_rate": 0.007788460842172551, + "loss": 0.0853, + "num_input_tokens_seen": 46512640, + "step": 35880 + }, + { + "epoch": 1.753377470500574, + "grad_norm": 0.0013360428856685758, + "learning_rate": 0.0077697376410706285, + "loss": 0.0732, + "num_input_tokens_seen": 46519296, + "step": 35885 + }, + { + "epoch": 1.7536217721642684, + "grad_norm": 0.0014471880858764052, + "learning_rate": 0.007751036373662567, + "loss": 0.0814, + "num_input_tokens_seen": 46525664, + "step": 35890 + }, + { + "epoch": 1.7538660738279628, + "grad_norm": 0.0015391232445836067, + "learning_rate": 0.00773235704283231, + "loss": 0.0763, + "num_input_tokens_seen": 46531968, + "step": 35895 + }, + { + "epoch": 1.7541103754916572, + "grad_norm": 0.0016289615305140615, + "learning_rate": 0.007713699651460437, + "loss": 0.0839, + "num_input_tokens_seen": 46538816, + "step": 35900 + }, + { + "epoch": 1.7543546771553515, + "grad_norm": 0.001153434976004064, + "learning_rate": 0.007695064202424162, + "loss": 0.0694, + "num_input_tokens_seen": 46545152, + "step": 35905 + }, + { + "epoch": 1.754598978819046, + "grad_norm": 0.0008419528021477163, + "learning_rate": 0.007676450698597286, + "loss": 0.0715, + "num_input_tokens_seen": 46551712, + "step": 35910 + }, + { + "epoch": 1.75484328048274, + "grad_norm": 0.0007018366595730186, + "learning_rate": 0.007657859142850265, + "loss": 0.0753, + "num_input_tokens_seen": 46558048, + "step": 35915 + }, + { + "epoch": 1.7550875821464345, + "grad_norm": 0.0015667134430259466, + "learning_rate": 0.0076392895380501535, + "loss": 0.0689, + "num_input_tokens_seen": 46564704, + "step": 35920 + }, + { + "epoch": 1.7553318838101286, + "grad_norm": 0.0014706762740388513, + "learning_rate": 0.007620741887060611, + "loss": 0.0788, + "num_input_tokens_seen": 46571232, + "step": 35925 + }, + { + "epoch": 1.755576185473823, + "grad_norm": 0.002881458029150963, + "learning_rate": 0.007602216192741901, + "loss": 0.1052, + "num_input_tokens_seen": 46577728, + "step": 35930 + }, + { + "epoch": 1.7558204871375174, + "grad_norm": 0.0007711853832006454, + "learning_rate": 0.007583712457950969, + "loss": 0.0648, + "num_input_tokens_seen": 46584096, + "step": 35935 + }, + { + "epoch": 1.7560647888012118, + "grad_norm": 0.0013780402950942516, + "learning_rate": 0.007565230685541269, + "loss": 0.1025, + "num_input_tokens_seen": 46590272, + "step": 35940 + }, + { + "epoch": 1.7563090904649061, + "grad_norm": 0.0009740476962178946, + "learning_rate": 0.007546770878362968, + "loss": 0.0799, + "num_input_tokens_seen": 46596320, + "step": 35945 + }, + { + "epoch": 1.7565533921286005, + "grad_norm": 0.001045404584147036, + "learning_rate": 0.0075283330392627405, + "loss": 0.0608, + "num_input_tokens_seen": 46603072, + "step": 35950 + }, + { + "epoch": 1.756797693792295, + "grad_norm": 0.0017572174547240138, + "learning_rate": 0.007509917171083979, + "loss": 0.0769, + "num_input_tokens_seen": 46610336, + "step": 35955 + }, + { + "epoch": 1.757041995455989, + "grad_norm": 0.0007804268971085548, + "learning_rate": 0.007491523276666662, + "loss": 0.0789, + "num_input_tokens_seen": 46616640, + "step": 35960 + }, + { + "epoch": 1.7572862971196834, + "grad_norm": 0.0010552761377766728, + "learning_rate": 0.007473151358847318, + "loss": 0.0829, + "num_input_tokens_seen": 46623232, + "step": 35965 + }, + { + "epoch": 1.7575305987833776, + "grad_norm": 0.001322745461948216, + "learning_rate": 0.007454801420459117, + "loss": 0.0803, + "num_input_tokens_seen": 46630112, + "step": 35970 + }, + { + "epoch": 1.757774900447072, + "grad_norm": 0.000732654589228332, + "learning_rate": 0.0074364734643319105, + "loss": 0.0606, + "num_input_tokens_seen": 46636800, + "step": 35975 + }, + { + "epoch": 1.7580192021107663, + "grad_norm": 0.0009708061115816236, + "learning_rate": 0.007418167493292022, + "loss": 0.0624, + "num_input_tokens_seen": 46643648, + "step": 35980 + }, + { + "epoch": 1.7582635037744607, + "grad_norm": 0.0016378203872591257, + "learning_rate": 0.0073998835101625245, + "loss": 0.0843, + "num_input_tokens_seen": 46649888, + "step": 35985 + }, + { + "epoch": 1.758507805438155, + "grad_norm": 0.0016375089762732387, + "learning_rate": 0.007381621517762998, + "loss": 0.0783, + "num_input_tokens_seen": 46656192, + "step": 35990 + }, + { + "epoch": 1.7587521071018495, + "grad_norm": 0.0011584263993427157, + "learning_rate": 0.007363381518909689, + "loss": 0.0854, + "num_input_tokens_seen": 46662784, + "step": 35995 + }, + { + "epoch": 1.7589964087655436, + "grad_norm": 0.0020841024816036224, + "learning_rate": 0.007345163516415448, + "loss": 0.0684, + "num_input_tokens_seen": 46669472, + "step": 36000 + }, + { + "epoch": 1.7589964087655436, + "eval_loss": 0.08395419269800186, + "eval_runtime": 402.221, + "eval_samples_per_second": 90.46, + "eval_steps_per_second": 22.617, + "num_input_tokens_seen": 46669472, + "step": 36000 + }, + { + "epoch": 1.759240710429238, + "grad_norm": 0.0012769944733008742, + "learning_rate": 0.007326967513089693, + "loss": 0.0826, + "num_input_tokens_seen": 46675936, + "step": 36005 + }, + { + "epoch": 1.7594850120929324, + "grad_norm": 0.0016073031583800912, + "learning_rate": 0.0073087935117384815, + "loss": 0.0868, + "num_input_tokens_seen": 46682272, + "step": 36010 + }, + { + "epoch": 1.7597293137566266, + "grad_norm": 0.0017044221749529243, + "learning_rate": 0.007290641515164503, + "loss": 0.0792, + "num_input_tokens_seen": 46688768, + "step": 36015 + }, + { + "epoch": 1.759973615420321, + "grad_norm": 0.0018006445607170463, + "learning_rate": 0.007272511526166986, + "loss": 0.0781, + "num_input_tokens_seen": 46695584, + "step": 36020 + }, + { + "epoch": 1.7602179170840153, + "grad_norm": 0.002102518454194069, + "learning_rate": 0.0072544035475418265, + "loss": 0.0861, + "num_input_tokens_seen": 46702144, + "step": 36025 + }, + { + "epoch": 1.7604622187477097, + "grad_norm": 0.0010804481571540236, + "learning_rate": 0.007236317582081475, + "loss": 0.062, + "num_input_tokens_seen": 46708672, + "step": 36030 + }, + { + "epoch": 1.760706520411404, + "grad_norm": 0.0012835422530770302, + "learning_rate": 0.007218253632575066, + "loss": 0.0994, + "num_input_tokens_seen": 46715008, + "step": 36035 + }, + { + "epoch": 1.7609508220750985, + "grad_norm": 0.0019867888186126947, + "learning_rate": 0.007200211701808223, + "loss": 0.0926, + "num_input_tokens_seen": 46721312, + "step": 36040 + }, + { + "epoch": 1.7611951237387926, + "grad_norm": 0.0013905307278037071, + "learning_rate": 0.007182191792563286, + "loss": 0.0612, + "num_input_tokens_seen": 46728224, + "step": 36045 + }, + { + "epoch": 1.761439425402487, + "grad_norm": 0.001450864365324378, + "learning_rate": 0.0071641939076191145, + "loss": 0.0648, + "num_input_tokens_seen": 46734400, + "step": 36050 + }, + { + "epoch": 1.7616837270661814, + "grad_norm": 0.0012206535320729017, + "learning_rate": 0.007146218049751257, + "loss": 0.0535, + "num_input_tokens_seen": 46741088, + "step": 36055 + }, + { + "epoch": 1.7619280287298755, + "grad_norm": 0.0016782128950580955, + "learning_rate": 0.0071282642217317775, + "loss": 0.0731, + "num_input_tokens_seen": 46747488, + "step": 36060 + }, + { + "epoch": 1.76217233039357, + "grad_norm": 0.0019031146075576544, + "learning_rate": 0.007110332426329396, + "loss": 0.0967, + "num_input_tokens_seen": 46753664, + "step": 36065 + }, + { + "epoch": 1.7624166320572643, + "grad_norm": 0.001260171877220273, + "learning_rate": 0.007092422666309417, + "loss": 0.073, + "num_input_tokens_seen": 46760544, + "step": 36070 + }, + { + "epoch": 1.7626609337209587, + "grad_norm": 0.0017408007988706231, + "learning_rate": 0.0070745349444337295, + "loss": 0.0894, + "num_input_tokens_seen": 46766880, + "step": 36075 + }, + { + "epoch": 1.762905235384653, + "grad_norm": 0.001176239107735455, + "learning_rate": 0.007056669263460913, + "loss": 0.0757, + "num_input_tokens_seen": 46773024, + "step": 36080 + }, + { + "epoch": 1.7631495370483474, + "grad_norm": 0.0005001295357942581, + "learning_rate": 0.007038825626145995, + "loss": 0.0706, + "num_input_tokens_seen": 46779744, + "step": 36085 + }, + { + "epoch": 1.7633938387120416, + "grad_norm": 0.0017883733380585909, + "learning_rate": 0.007021004035240724, + "loss": 0.083, + "num_input_tokens_seen": 46786304, + "step": 36090 + }, + { + "epoch": 1.763638140375736, + "grad_norm": 0.0012796991504728794, + "learning_rate": 0.007003204493493453, + "loss": 0.0985, + "num_input_tokens_seen": 46792896, + "step": 36095 + }, + { + "epoch": 1.7638824420394303, + "grad_norm": 0.0010613021440804005, + "learning_rate": 0.006985427003649036, + "loss": 0.0735, + "num_input_tokens_seen": 46799232, + "step": 36100 + }, + { + "epoch": 1.7641267437031245, + "grad_norm": 0.0007555551128461957, + "learning_rate": 0.006967671568449013, + "loss": 0.077, + "num_input_tokens_seen": 46805856, + "step": 36105 + }, + { + "epoch": 1.7643710453668189, + "grad_norm": 0.0010948276612907648, + "learning_rate": 0.006949938190631511, + "loss": 0.0693, + "num_input_tokens_seen": 46812512, + "step": 36110 + }, + { + "epoch": 1.7646153470305133, + "grad_norm": 0.0019063956569880247, + "learning_rate": 0.0069322268729311905, + "loss": 0.0642, + "num_input_tokens_seen": 46819968, + "step": 36115 + }, + { + "epoch": 1.7648596486942076, + "grad_norm": 0.001835654373280704, + "learning_rate": 0.006914537618079403, + "loss": 0.0869, + "num_input_tokens_seen": 46826080, + "step": 36120 + }, + { + "epoch": 1.765103950357902, + "grad_norm": 0.0031655202619731426, + "learning_rate": 0.006896870428804031, + "loss": 0.0879, + "num_input_tokens_seen": 46832896, + "step": 36125 + }, + { + "epoch": 1.7653482520215964, + "grad_norm": 0.0009861462749540806, + "learning_rate": 0.006879225307829595, + "loss": 0.0796, + "num_input_tokens_seen": 46839552, + "step": 36130 + }, + { + "epoch": 1.7655925536852906, + "grad_norm": 0.0011173608945682645, + "learning_rate": 0.00686160225787717, + "loss": 0.0654, + "num_input_tokens_seen": 46845568, + "step": 36135 + }, + { + "epoch": 1.765836855348985, + "grad_norm": 0.0007200298132374883, + "learning_rate": 0.006844001281664463, + "loss": 0.0467, + "num_input_tokens_seen": 46852416, + "step": 36140 + }, + { + "epoch": 1.766081157012679, + "grad_norm": 0.001096520689316094, + "learning_rate": 0.006826422381905789, + "loss": 0.0687, + "num_input_tokens_seen": 46859008, + "step": 36145 + }, + { + "epoch": 1.7663254586763735, + "grad_norm": 0.0013321596197783947, + "learning_rate": 0.006808865561311994, + "loss": 0.0952, + "num_input_tokens_seen": 46865248, + "step": 36150 + }, + { + "epoch": 1.7665697603400679, + "grad_norm": 0.002596938516944647, + "learning_rate": 0.00679133082259058, + "loss": 0.1032, + "num_input_tokens_seen": 46871872, + "step": 36155 + }, + { + "epoch": 1.7668140620037622, + "grad_norm": 0.002128775930032134, + "learning_rate": 0.00677381816844565, + "loss": 0.1057, + "num_input_tokens_seen": 46878208, + "step": 36160 + }, + { + "epoch": 1.7670583636674566, + "grad_norm": 0.0016743146115913987, + "learning_rate": 0.0067563276015778434, + "loss": 0.08, + "num_input_tokens_seen": 46884864, + "step": 36165 + }, + { + "epoch": 1.767302665331151, + "grad_norm": 0.0016406795475631952, + "learning_rate": 0.006738859124684437, + "loss": 0.0624, + "num_input_tokens_seen": 46891360, + "step": 36170 + }, + { + "epoch": 1.7675469669948454, + "grad_norm": 0.002097462071105838, + "learning_rate": 0.006721412740459259, + "loss": 0.0766, + "num_input_tokens_seen": 46897344, + "step": 36175 + }, + { + "epoch": 1.7677912686585395, + "grad_norm": 0.0027752756141126156, + "learning_rate": 0.006703988451592824, + "loss": 0.0739, + "num_input_tokens_seen": 46903488, + "step": 36180 + }, + { + "epoch": 1.768035570322234, + "grad_norm": 0.0011213211109861732, + "learning_rate": 0.006686586260772114, + "loss": 0.0691, + "num_input_tokens_seen": 46909984, + "step": 36185 + }, + { + "epoch": 1.768279871985928, + "grad_norm": 0.0008743772050365806, + "learning_rate": 0.006669206170680819, + "loss": 0.09, + "num_input_tokens_seen": 46916736, + "step": 36190 + }, + { + "epoch": 1.7685241736496224, + "grad_norm": 0.0011327799875289202, + "learning_rate": 0.0066518481839991095, + "loss": 0.08, + "num_input_tokens_seen": 46923072, + "step": 36195 + }, + { + "epoch": 1.7687684753133168, + "grad_norm": 0.0011758811306208372, + "learning_rate": 0.006634512303403861, + "loss": 0.0909, + "num_input_tokens_seen": 46929280, + "step": 36200 + }, + { + "epoch": 1.7687684753133168, + "eval_loss": 0.08395708352327347, + "eval_runtime": 402.1654, + "eval_samples_per_second": 90.473, + "eval_steps_per_second": 22.62, + "num_input_tokens_seen": 46929280, + "step": 36200 + }, + { + "epoch": 1.7690127769770112, + "grad_norm": 0.0014215223491191864, + "learning_rate": 0.0066171985315684355, + "loss": 0.0728, + "num_input_tokens_seen": 46935392, + "step": 36205 + }, + { + "epoch": 1.7692570786407056, + "grad_norm": 0.00178151682484895, + "learning_rate": 0.0065999068711628806, + "loss": 0.0728, + "num_input_tokens_seen": 46942176, + "step": 36210 + }, + { + "epoch": 1.7695013803044, + "grad_norm": 0.0012473227689042687, + "learning_rate": 0.0065826373248537295, + "loss": 0.0883, + "num_input_tokens_seen": 46948960, + "step": 36215 + }, + { + "epoch": 1.7697456819680943, + "grad_norm": 0.000970689405221492, + "learning_rate": 0.006565389895304218, + "loss": 0.0845, + "num_input_tokens_seen": 46955712, + "step": 36220 + }, + { + "epoch": 1.7699899836317885, + "grad_norm": 0.001678053173236549, + "learning_rate": 0.006548164585174104, + "loss": 0.0734, + "num_input_tokens_seen": 46962208, + "step": 36225 + }, + { + "epoch": 1.7702342852954829, + "grad_norm": 0.0009457849082536995, + "learning_rate": 0.006530961397119728, + "loss": 0.0825, + "num_input_tokens_seen": 46968672, + "step": 36230 + }, + { + "epoch": 1.770478586959177, + "grad_norm": 0.001334735774435103, + "learning_rate": 0.00651378033379405, + "loss": 0.0785, + "num_input_tokens_seen": 46975264, + "step": 36235 + }, + { + "epoch": 1.7707228886228714, + "grad_norm": 0.0008355446043424308, + "learning_rate": 0.006496621397846619, + "loss": 0.0502, + "num_input_tokens_seen": 46981728, + "step": 36240 + }, + { + "epoch": 1.7709671902865658, + "grad_norm": 0.00172431580722332, + "learning_rate": 0.006479484591923518, + "loss": 0.0801, + "num_input_tokens_seen": 46988288, + "step": 36245 + }, + { + "epoch": 1.7712114919502602, + "grad_norm": 0.0016464685322716832, + "learning_rate": 0.006462369918667515, + "loss": 0.0703, + "num_input_tokens_seen": 46995296, + "step": 36250 + }, + { + "epoch": 1.7714557936139546, + "grad_norm": 0.0018083214526996017, + "learning_rate": 0.006445277380717851, + "loss": 0.0899, + "num_input_tokens_seen": 47001664, + "step": 36255 + }, + { + "epoch": 1.771700095277649, + "grad_norm": 0.0011571487411856651, + "learning_rate": 0.006428206980710466, + "loss": 0.0973, + "num_input_tokens_seen": 47007904, + "step": 36260 + }, + { + "epoch": 1.7719443969413433, + "grad_norm": 0.0016499857883900404, + "learning_rate": 0.006411158721277788, + "loss": 0.0953, + "num_input_tokens_seen": 47013696, + "step": 36265 + }, + { + "epoch": 1.7721886986050375, + "grad_norm": 0.00257277674973011, + "learning_rate": 0.00639413260504888, + "loss": 0.097, + "num_input_tokens_seen": 47020352, + "step": 36270 + }, + { + "epoch": 1.7724330002687319, + "grad_norm": 0.0007466040551662445, + "learning_rate": 0.006377128634649376, + "loss": 0.065, + "num_input_tokens_seen": 47026592, + "step": 36275 + }, + { + "epoch": 1.772677301932426, + "grad_norm": 0.001176884863525629, + "learning_rate": 0.006360146812701528, + "loss": 0.0619, + "num_input_tokens_seen": 47032896, + "step": 36280 + }, + { + "epoch": 1.7729216035961204, + "grad_norm": 0.0010602507973089814, + "learning_rate": 0.006343187141824125, + "loss": 0.0663, + "num_input_tokens_seen": 47040192, + "step": 36285 + }, + { + "epoch": 1.7731659052598148, + "grad_norm": 0.001362764509394765, + "learning_rate": 0.00632624962463259, + "loss": 0.0658, + "num_input_tokens_seen": 47046752, + "step": 36290 + }, + { + "epoch": 1.7734102069235091, + "grad_norm": 0.0013637141091749072, + "learning_rate": 0.006309334263738853, + "loss": 0.0734, + "num_input_tokens_seen": 47053472, + "step": 36295 + }, + { + "epoch": 1.7736545085872035, + "grad_norm": 0.0016276014503091574, + "learning_rate": 0.006292441061751508, + "loss": 0.0728, + "num_input_tokens_seen": 47059776, + "step": 36300 + }, + { + "epoch": 1.773898810250898, + "grad_norm": 0.001249409862793982, + "learning_rate": 0.0062755700212757054, + "loss": 0.0674, + "num_input_tokens_seen": 47066048, + "step": 36305 + }, + { + "epoch": 1.7741431119145923, + "grad_norm": 0.001362260547466576, + "learning_rate": 0.006258721144913148, + "loss": 0.0612, + "num_input_tokens_seen": 47072512, + "step": 36310 + }, + { + "epoch": 1.7743874135782864, + "grad_norm": 0.002001541666686535, + "learning_rate": 0.0062418944352621575, + "loss": 0.0975, + "num_input_tokens_seen": 47078816, + "step": 36315 + }, + { + "epoch": 1.7746317152419808, + "grad_norm": 0.0017799480119720101, + "learning_rate": 0.0062250898949176405, + "loss": 0.0739, + "num_input_tokens_seen": 47085216, + "step": 36320 + }, + { + "epoch": 1.774876016905675, + "grad_norm": 0.0017213402315974236, + "learning_rate": 0.006208307526471041, + "loss": 0.0874, + "num_input_tokens_seen": 47091488, + "step": 36325 + }, + { + "epoch": 1.7751203185693694, + "grad_norm": 0.0019346224144101143, + "learning_rate": 0.006191547332510405, + "loss": 0.1046, + "num_input_tokens_seen": 47097920, + "step": 36330 + }, + { + "epoch": 1.7753646202330637, + "grad_norm": 0.0012710614828392863, + "learning_rate": 0.006174809315620416, + "loss": 0.0822, + "num_input_tokens_seen": 47104288, + "step": 36335 + }, + { + "epoch": 1.7756089218967581, + "grad_norm": 0.001268469961360097, + "learning_rate": 0.00615809347838221, + "loss": 0.0746, + "num_input_tokens_seen": 47110976, + "step": 36340 + }, + { + "epoch": 1.7758532235604525, + "grad_norm": 0.001128287985920906, + "learning_rate": 0.006141399823373655, + "loss": 0.0729, + "num_input_tokens_seen": 47117568, + "step": 36345 + }, + { + "epoch": 1.7760975252241469, + "grad_norm": 0.0019609450828284025, + "learning_rate": 0.0061247283531690455, + "loss": 0.0872, + "num_input_tokens_seen": 47123712, + "step": 36350 + }, + { + "epoch": 1.7763418268878413, + "grad_norm": 0.0016167059075087309, + "learning_rate": 0.0061080790703393895, + "loss": 0.1036, + "num_input_tokens_seen": 47130208, + "step": 36355 + }, + { + "epoch": 1.7765861285515354, + "grad_norm": 0.0017174076056107879, + "learning_rate": 0.006091451977452217, + "loss": 0.0789, + "num_input_tokens_seen": 47136992, + "step": 36360 + }, + { + "epoch": 1.7768304302152298, + "grad_norm": 0.0020616105757653713, + "learning_rate": 0.00607484707707161, + "loss": 0.0626, + "num_input_tokens_seen": 47143712, + "step": 36365 + }, + { + "epoch": 1.777074731878924, + "grad_norm": 0.0013858586316928267, + "learning_rate": 0.006058264371758254, + "loss": 0.0875, + "num_input_tokens_seen": 47150144, + "step": 36370 + }, + { + "epoch": 1.7773190335426183, + "grad_norm": 0.0012409210903570056, + "learning_rate": 0.00604170386406942, + "loss": 0.0876, + "num_input_tokens_seen": 47156192, + "step": 36375 + }, + { + "epoch": 1.7775633352063127, + "grad_norm": 0.0015198991168290377, + "learning_rate": 0.006025165556558931, + "loss": 0.1193, + "num_input_tokens_seen": 47162560, + "step": 36380 + }, + { + "epoch": 1.777807636870007, + "grad_norm": 0.0017000283114612103, + "learning_rate": 0.006008649451777248, + "loss": 0.0993, + "num_input_tokens_seen": 47168544, + "step": 36385 + }, + { + "epoch": 1.7780519385337015, + "grad_norm": 0.002553916536271572, + "learning_rate": 0.005992155552271283, + "loss": 0.0876, + "num_input_tokens_seen": 47174848, + "step": 36390 + }, + { + "epoch": 1.7782962401973958, + "grad_norm": 0.0018588491948321462, + "learning_rate": 0.005975683860584685, + "loss": 0.1232, + "num_input_tokens_seen": 47181536, + "step": 36395 + }, + { + "epoch": 1.7785405418610902, + "grad_norm": 0.0012795397778972983, + "learning_rate": 0.0059592343792575385, + "loss": 0.1004, + "num_input_tokens_seen": 47188416, + "step": 36400 + }, + { + "epoch": 1.7785405418610902, + "eval_loss": 0.08354517817497253, + "eval_runtime": 402.2437, + "eval_samples_per_second": 90.455, + "eval_steps_per_second": 22.616, + "num_input_tokens_seen": 47188416, + "step": 36400 + }, + { + "epoch": 1.7787848435247844, + "grad_norm": 0.0015459326095879078, + "learning_rate": 0.0059428071108265975, + "loss": 0.0714, + "num_input_tokens_seen": 47194688, + "step": 36405 + }, + { + "epoch": 1.7790291451884788, + "grad_norm": 0.0016670970944687724, + "learning_rate": 0.005926402057825136, + "loss": 0.0844, + "num_input_tokens_seen": 47201216, + "step": 36410 + }, + { + "epoch": 1.779273446852173, + "grad_norm": 0.0015203492948785424, + "learning_rate": 0.005910019222782997, + "loss": 0.0666, + "num_input_tokens_seen": 47207744, + "step": 36415 + }, + { + "epoch": 1.7795177485158673, + "grad_norm": 0.0012730122543871403, + "learning_rate": 0.005893658608226643, + "loss": 0.0763, + "num_input_tokens_seen": 47213984, + "step": 36420 + }, + { + "epoch": 1.7797620501795617, + "grad_norm": 0.0009145904914475977, + "learning_rate": 0.0058773202166791045, + "loss": 0.0875, + "num_input_tokens_seen": 47220320, + "step": 36425 + }, + { + "epoch": 1.780006351843256, + "grad_norm": 0.0014221048913896084, + "learning_rate": 0.005861004050659918, + "loss": 0.0709, + "num_input_tokens_seen": 47226592, + "step": 36430 + }, + { + "epoch": 1.7802506535069504, + "grad_norm": 0.0012850328348577023, + "learning_rate": 0.005844710112685286, + "loss": 0.0837, + "num_input_tokens_seen": 47233056, + "step": 36435 + }, + { + "epoch": 1.7804949551706448, + "grad_norm": 0.001905736862681806, + "learning_rate": 0.005828438405267933, + "loss": 0.0793, + "num_input_tokens_seen": 47239776, + "step": 36440 + }, + { + "epoch": 1.7807392568343392, + "grad_norm": 0.0007037604227662086, + "learning_rate": 0.00581218893091715, + "loss": 0.073, + "num_input_tokens_seen": 47246080, + "step": 36445 + }, + { + "epoch": 1.7809835584980334, + "grad_norm": 0.0014800102217122912, + "learning_rate": 0.005795961692138801, + "loss": 0.0904, + "num_input_tokens_seen": 47252512, + "step": 36450 + }, + { + "epoch": 1.7812278601617277, + "grad_norm": 0.0010102033847942948, + "learning_rate": 0.00577975669143535, + "loss": 0.1005, + "num_input_tokens_seen": 47259136, + "step": 36455 + }, + { + "epoch": 1.781472161825422, + "grad_norm": 0.0010939744533970952, + "learning_rate": 0.005763573931305782, + "loss": 0.0986, + "num_input_tokens_seen": 47265632, + "step": 36460 + }, + { + "epoch": 1.7817164634891163, + "grad_norm": 0.0018313798354938626, + "learning_rate": 0.005747413414245733, + "loss": 0.0664, + "num_input_tokens_seen": 47272000, + "step": 36465 + }, + { + "epoch": 1.7819607651528107, + "grad_norm": 0.0008164487662725151, + "learning_rate": 0.005731275142747294, + "loss": 0.066, + "num_input_tokens_seen": 47278272, + "step": 36470 + }, + { + "epoch": 1.782205066816505, + "grad_norm": 0.0007896117167547345, + "learning_rate": 0.005715159119299256, + "loss": 0.0552, + "num_input_tokens_seen": 47284928, + "step": 36475 + }, + { + "epoch": 1.7824493684801994, + "grad_norm": 0.0012635414022952318, + "learning_rate": 0.005699065346386867, + "loss": 0.0553, + "num_input_tokens_seen": 47291328, + "step": 36480 + }, + { + "epoch": 1.7826936701438938, + "grad_norm": 0.0016963215311989188, + "learning_rate": 0.0056829938264919885, + "loss": 0.0722, + "num_input_tokens_seen": 47298272, + "step": 36485 + }, + { + "epoch": 1.7829379718075882, + "grad_norm": 0.0011983716394752264, + "learning_rate": 0.005666944562093074, + "loss": 0.0672, + "num_input_tokens_seen": 47304736, + "step": 36490 + }, + { + "epoch": 1.7831822734712823, + "grad_norm": 0.0013005806831642985, + "learning_rate": 0.005650917555665108, + "loss": 0.0991, + "num_input_tokens_seen": 47310976, + "step": 36495 + }, + { + "epoch": 1.7834265751349767, + "grad_norm": 0.0019359763246029615, + "learning_rate": 0.005634912809679632, + "loss": 0.0749, + "num_input_tokens_seen": 47317696, + "step": 36500 + }, + { + "epoch": 1.7836708767986709, + "grad_norm": 0.0016266792081296444, + "learning_rate": 0.005618930326604854, + "loss": 0.0903, + "num_input_tokens_seen": 47323552, + "step": 36505 + }, + { + "epoch": 1.7839151784623652, + "grad_norm": 0.0013981560477986932, + "learning_rate": 0.005602970108905386, + "loss": 0.099, + "num_input_tokens_seen": 47330208, + "step": 36510 + }, + { + "epoch": 1.7841594801260596, + "grad_norm": 0.0015954438131302595, + "learning_rate": 0.005587032159042543, + "loss": 0.0703, + "num_input_tokens_seen": 47337088, + "step": 36515 + }, + { + "epoch": 1.784403781789754, + "grad_norm": 0.0013201572000980377, + "learning_rate": 0.005571116479474158, + "loss": 0.0675, + "num_input_tokens_seen": 47343712, + "step": 36520 + }, + { + "epoch": 1.7846480834534484, + "grad_norm": 0.001112280529923737, + "learning_rate": 0.005555223072654619, + "loss": 0.0574, + "num_input_tokens_seen": 47350240, + "step": 36525 + }, + { + "epoch": 1.7848923851171428, + "grad_norm": 0.0008494507637806237, + "learning_rate": 0.005539351941034881, + "loss": 0.0604, + "num_input_tokens_seen": 47357248, + "step": 36530 + }, + { + "epoch": 1.785136686780837, + "grad_norm": 0.0011150408536195755, + "learning_rate": 0.0055235030870624865, + "loss": 0.0634, + "num_input_tokens_seen": 47363712, + "step": 36535 + }, + { + "epoch": 1.7853809884445313, + "grad_norm": 0.0009130594553425908, + "learning_rate": 0.005507676513181514, + "loss": 0.068, + "num_input_tokens_seen": 47370336, + "step": 36540 + }, + { + "epoch": 1.7856252901082257, + "grad_norm": 0.0010137163335457444, + "learning_rate": 0.005491872221832628, + "loss": 0.0646, + "num_input_tokens_seen": 47376800, + "step": 36545 + }, + { + "epoch": 1.7858695917719198, + "grad_norm": 0.000970973982475698, + "learning_rate": 0.005476090215453061, + "loss": 0.0594, + "num_input_tokens_seen": 47383360, + "step": 36550 + }, + { + "epoch": 1.7861138934356142, + "grad_norm": 0.0008891918114386499, + "learning_rate": 0.0054603304964765675, + "loss": 0.0914, + "num_input_tokens_seen": 47389696, + "step": 36555 + }, + { + "epoch": 1.7863581950993086, + "grad_norm": 0.0013394871493801475, + "learning_rate": 0.005444593067333519, + "loss": 0.0654, + "num_input_tokens_seen": 47396128, + "step": 36560 + }, + { + "epoch": 1.786602496763003, + "grad_norm": 0.0009942421456798911, + "learning_rate": 0.00542887793045081, + "loss": 0.0697, + "num_input_tokens_seen": 47402880, + "step": 36565 + }, + { + "epoch": 1.7868467984266974, + "grad_norm": 0.0016244301805272698, + "learning_rate": 0.005413185088251932, + "loss": 0.082, + "num_input_tokens_seen": 47409376, + "step": 36570 + }, + { + "epoch": 1.7870911000903917, + "grad_norm": 0.0014899030793458223, + "learning_rate": 0.005397514543156884, + "loss": 0.0734, + "num_input_tokens_seen": 47415520, + "step": 36575 + }, + { + "epoch": 1.787335401754086, + "grad_norm": 0.0013459959300234914, + "learning_rate": 0.0053818662975822825, + "loss": 0.0788, + "num_input_tokens_seen": 47421984, + "step": 36580 + }, + { + "epoch": 1.7875797034177803, + "grad_norm": 0.0024221723433583975, + "learning_rate": 0.005366240353941315, + "loss": 0.0744, + "num_input_tokens_seen": 47428480, + "step": 36585 + }, + { + "epoch": 1.7878240050814747, + "grad_norm": 0.001613355241715908, + "learning_rate": 0.005350636714643636, + "loss": 0.0778, + "num_input_tokens_seen": 47434944, + "step": 36590 + }, + { + "epoch": 1.7880683067451688, + "grad_norm": 0.0016721680294722319, + "learning_rate": 0.005335055382095555, + "loss": 0.0654, + "num_input_tokens_seen": 47441088, + "step": 36595 + }, + { + "epoch": 1.7883126084088632, + "grad_norm": 0.00182488898281008, + "learning_rate": 0.005319496358699915, + "loss": 0.0782, + "num_input_tokens_seen": 47447328, + "step": 36600 + }, + { + "epoch": 1.7883126084088632, + "eval_loss": 0.08366671204566956, + "eval_runtime": 402.2391, + "eval_samples_per_second": 90.456, + "eval_steps_per_second": 22.616, + "num_input_tokens_seen": 47447328, + "step": 36600 + }, + { + "epoch": 1.7885569100725576, + "grad_norm": 0.0012292176252231002, + "learning_rate": 0.005303959646856099, + "loss": 0.0871, + "num_input_tokens_seen": 47454016, + "step": 36605 + }, + { + "epoch": 1.788801211736252, + "grad_norm": 0.0017138496041297913, + "learning_rate": 0.005288445248960089, + "loss": 0.075, + "num_input_tokens_seen": 47460576, + "step": 36610 + }, + { + "epoch": 1.7890455133999463, + "grad_norm": 0.0012183758663013577, + "learning_rate": 0.005272953167404354, + "loss": 0.0792, + "num_input_tokens_seen": 47467040, + "step": 36615 + }, + { + "epoch": 1.7892898150636407, + "grad_norm": 0.001916942186653614, + "learning_rate": 0.005257483404578017, + "loss": 0.0763, + "num_input_tokens_seen": 47473408, + "step": 36620 + }, + { + "epoch": 1.7895341167273349, + "grad_norm": 0.0011407637502998114, + "learning_rate": 0.0052420359628666865, + "loss": 0.0658, + "num_input_tokens_seen": 47479872, + "step": 36625 + }, + { + "epoch": 1.7897784183910292, + "grad_norm": 0.001800475176423788, + "learning_rate": 0.00522661084465254, + "loss": 0.0925, + "num_input_tokens_seen": 47486368, + "step": 36630 + }, + { + "epoch": 1.7900227200547236, + "grad_norm": 0.001238205237314105, + "learning_rate": 0.005211208052314326, + "loss": 0.0933, + "num_input_tokens_seen": 47492832, + "step": 36635 + }, + { + "epoch": 1.7902670217184178, + "grad_norm": 0.0014189665671437979, + "learning_rate": 0.005195827588227391, + "loss": 0.0771, + "num_input_tokens_seen": 47499360, + "step": 36640 + }, + { + "epoch": 1.7905113233821122, + "grad_norm": 0.001257703872397542, + "learning_rate": 0.0051804694547635255, + "loss": 0.0922, + "num_input_tokens_seen": 47506016, + "step": 36645 + }, + { + "epoch": 1.7907556250458065, + "grad_norm": 0.001807679422199726, + "learning_rate": 0.005165133654291232, + "loss": 0.076, + "num_input_tokens_seen": 47512160, + "step": 36650 + }, + { + "epoch": 1.790999926709501, + "grad_norm": 0.0007054567686282098, + "learning_rate": 0.005149820189175402, + "loss": 0.0523, + "num_input_tokens_seen": 47519072, + "step": 36655 + }, + { + "epoch": 1.7912442283731953, + "grad_norm": 0.0017423543613404036, + "learning_rate": 0.005134529061777598, + "loss": 0.0869, + "num_input_tokens_seen": 47525632, + "step": 36660 + }, + { + "epoch": 1.7914885300368897, + "grad_norm": 0.0010766396299004555, + "learning_rate": 0.005119260274455933, + "loss": 0.0713, + "num_input_tokens_seen": 47531520, + "step": 36665 + }, + { + "epoch": 1.7917328317005838, + "grad_norm": 0.0021267500706017017, + "learning_rate": 0.005104013829565007, + "loss": 0.1053, + "num_input_tokens_seen": 47537760, + "step": 36670 + }, + { + "epoch": 1.7919771333642782, + "grad_norm": 0.0006279356894083321, + "learning_rate": 0.005088789729456006, + "loss": 0.0527, + "num_input_tokens_seen": 47544448, + "step": 36675 + }, + { + "epoch": 1.7922214350279724, + "grad_norm": 0.0025097664911299944, + "learning_rate": 0.005073587976476735, + "loss": 0.1115, + "num_input_tokens_seen": 47550944, + "step": 36680 + }, + { + "epoch": 1.7924657366916668, + "grad_norm": 0.001554017886519432, + "learning_rate": 0.005058408572971418, + "loss": 0.0775, + "num_input_tokens_seen": 47556864, + "step": 36685 + }, + { + "epoch": 1.7927100383553611, + "grad_norm": 0.0013238782994449139, + "learning_rate": 0.005043251521280983, + "loss": 0.0651, + "num_input_tokens_seen": 47563616, + "step": 36690 + }, + { + "epoch": 1.7929543400190555, + "grad_norm": 0.0015538616571575403, + "learning_rate": 0.005028116823742795, + "loss": 0.0743, + "num_input_tokens_seen": 47570304, + "step": 36695 + }, + { + "epoch": 1.79319864168275, + "grad_norm": 0.0010407809168100357, + "learning_rate": 0.005013004482690819, + "loss": 0.0582, + "num_input_tokens_seen": 47576896, + "step": 36700 + }, + { + "epoch": 1.7934429433464443, + "grad_norm": 0.0013279292033985257, + "learning_rate": 0.0049979145004555746, + "loss": 0.0809, + "num_input_tokens_seen": 47583680, + "step": 36705 + }, + { + "epoch": 1.7936872450101387, + "grad_norm": 0.000890987750608474, + "learning_rate": 0.004982846879364116, + "loss": 0.0703, + "num_input_tokens_seen": 47590368, + "step": 36710 + }, + { + "epoch": 1.7939315466738328, + "grad_norm": 0.0010336264967918396, + "learning_rate": 0.0049678016217400535, + "loss": 0.0736, + "num_input_tokens_seen": 47596512, + "step": 36715 + }, + { + "epoch": 1.7941758483375272, + "grad_norm": 0.0013811668613925576, + "learning_rate": 0.004952778729903595, + "loss": 0.0946, + "num_input_tokens_seen": 47602784, + "step": 36720 + }, + { + "epoch": 1.7944201500012213, + "grad_norm": 0.0012859581038355827, + "learning_rate": 0.004937778206171422, + "loss": 0.0995, + "num_input_tokens_seen": 47608960, + "step": 36725 + }, + { + "epoch": 1.7946644516649157, + "grad_norm": 0.0011467199074104428, + "learning_rate": 0.004922800052856835, + "loss": 0.0612, + "num_input_tokens_seen": 47615456, + "step": 36730 + }, + { + "epoch": 1.79490875332861, + "grad_norm": 0.0006828680052421987, + "learning_rate": 0.004907844272269602, + "loss": 0.072, + "num_input_tokens_seen": 47621856, + "step": 36735 + }, + { + "epoch": 1.7951530549923045, + "grad_norm": 0.0010589630110189319, + "learning_rate": 0.004892910866716144, + "loss": 0.0938, + "num_input_tokens_seen": 47627904, + "step": 36740 + }, + { + "epoch": 1.7953973566559989, + "grad_norm": 0.0009505319758318365, + "learning_rate": 0.004877999838499369, + "loss": 0.0621, + "num_input_tokens_seen": 47635008, + "step": 36745 + }, + { + "epoch": 1.7956416583196932, + "grad_norm": 0.0011341844219714403, + "learning_rate": 0.0048631111899187065, + "loss": 0.0749, + "num_input_tokens_seen": 47641344, + "step": 36750 + }, + { + "epoch": 1.7958859599833876, + "grad_norm": 0.0013535887701436877, + "learning_rate": 0.0048482449232702335, + "loss": 0.1006, + "num_input_tokens_seen": 47648064, + "step": 36755 + }, + { + "epoch": 1.7961302616470818, + "grad_norm": 0.001723959343507886, + "learning_rate": 0.004833401040846469, + "loss": 0.0821, + "num_input_tokens_seen": 47654816, + "step": 36760 + }, + { + "epoch": 1.7963745633107762, + "grad_norm": 0.0024823681451380253, + "learning_rate": 0.004818579544936546, + "loss": 0.1015, + "num_input_tokens_seen": 47661440, + "step": 36765 + }, + { + "epoch": 1.7966188649744703, + "grad_norm": 0.0012037154519930482, + "learning_rate": 0.004803780437826121, + "loss": 0.0766, + "num_input_tokens_seen": 47667776, + "step": 36770 + }, + { + "epoch": 1.7968631666381647, + "grad_norm": 0.0007028841646388173, + "learning_rate": 0.004789003721797402, + "loss": 0.0475, + "num_input_tokens_seen": 47674080, + "step": 36775 + }, + { + "epoch": 1.797107468301859, + "grad_norm": 0.0010387094225734472, + "learning_rate": 0.004774249399129132, + "loss": 0.0745, + "num_input_tokens_seen": 47680416, + "step": 36780 + }, + { + "epoch": 1.7973517699655535, + "grad_norm": 0.0019055560696870089, + "learning_rate": 0.004759517472096642, + "loss": 0.0929, + "num_input_tokens_seen": 47687328, + "step": 36785 + }, + { + "epoch": 1.7975960716292478, + "grad_norm": 0.001041870447807014, + "learning_rate": 0.004744807942971746, + "loss": 0.0609, + "num_input_tokens_seen": 47693696, + "step": 36790 + }, + { + "epoch": 1.7978403732929422, + "grad_norm": 0.0009920793818309903, + "learning_rate": 0.004730120814022881, + "loss": 0.0941, + "num_input_tokens_seen": 47700128, + "step": 36795 + }, + { + "epoch": 1.7980846749566366, + "grad_norm": 0.001157553750090301, + "learning_rate": 0.004715456087514935, + "loss": 0.0857, + "num_input_tokens_seen": 47707040, + "step": 36800 + }, + { + "epoch": 1.7980846749566366, + "eval_loss": 0.08350171893835068, + "eval_runtime": 402.5287, + "eval_samples_per_second": 90.391, + "eval_steps_per_second": 22.6, + "num_input_tokens_seen": 47707040, + "step": 36800 + }, + { + "epoch": 1.7983289766203308, + "grad_norm": 0.0027302305679768324, + "learning_rate": 0.004700813765709432, + "loss": 0.083, + "num_input_tokens_seen": 47713664, + "step": 36805 + }, + { + "epoch": 1.7985732782840251, + "grad_norm": 0.0008788372506387532, + "learning_rate": 0.004686193850864401, + "loss": 0.0626, + "num_input_tokens_seen": 47720416, + "step": 36810 + }, + { + "epoch": 1.7988175799477193, + "grad_norm": 0.0012569454265758395, + "learning_rate": 0.004671596345234385, + "loss": 0.0922, + "num_input_tokens_seen": 47727040, + "step": 36815 + }, + { + "epoch": 1.7990618816114137, + "grad_norm": 0.0020337793976068497, + "learning_rate": 0.00465702125107052, + "loss": 0.0901, + "num_input_tokens_seen": 47733952, + "step": 36820 + }, + { + "epoch": 1.799306183275108, + "grad_norm": 0.0007955830078572035, + "learning_rate": 0.004642468570620506, + "loss": 0.0485, + "num_input_tokens_seen": 47740512, + "step": 36825 + }, + { + "epoch": 1.7995504849388024, + "grad_norm": 0.0014713736018165946, + "learning_rate": 0.004627938306128482, + "loss": 0.0669, + "num_input_tokens_seen": 47747008, + "step": 36830 + }, + { + "epoch": 1.7997947866024968, + "grad_norm": 0.000737716443836689, + "learning_rate": 0.004613430459835255, + "loss": 0.0501, + "num_input_tokens_seen": 47753856, + "step": 36835 + }, + { + "epoch": 1.8000390882661912, + "grad_norm": 0.0017120889388024807, + "learning_rate": 0.004598945033978085, + "loss": 0.0789, + "num_input_tokens_seen": 47760192, + "step": 36840 + }, + { + "epoch": 1.8002833899298856, + "grad_norm": 0.0012573780259117484, + "learning_rate": 0.004584482030790804, + "loss": 0.0739, + "num_input_tokens_seen": 47766976, + "step": 36845 + }, + { + "epoch": 1.8005276915935797, + "grad_norm": 0.0012021228903904557, + "learning_rate": 0.004570041452503826, + "loss": 0.0612, + "num_input_tokens_seen": 47773824, + "step": 36850 + }, + { + "epoch": 1.800771993257274, + "grad_norm": 0.000961262034252286, + "learning_rate": 0.004555623301344003, + "loss": 0.0652, + "num_input_tokens_seen": 47780448, + "step": 36855 + }, + { + "epoch": 1.8010162949209683, + "grad_norm": 0.0006132288253866136, + "learning_rate": 0.004541227579534857, + "loss": 0.0733, + "num_input_tokens_seen": 47786944, + "step": 36860 + }, + { + "epoch": 1.8012605965846626, + "grad_norm": 0.0011022869730368257, + "learning_rate": 0.004526854289296378, + "loss": 0.0566, + "num_input_tokens_seen": 47793344, + "step": 36865 + }, + { + "epoch": 1.801504898248357, + "grad_norm": 0.0009821447310969234, + "learning_rate": 0.004512503432845078, + "loss": 0.0695, + "num_input_tokens_seen": 47799904, + "step": 36870 + }, + { + "epoch": 1.8017491999120514, + "grad_norm": 0.0010867133969441056, + "learning_rate": 0.004498175012394068, + "loss": 0.0652, + "num_input_tokens_seen": 47806752, + "step": 36875 + }, + { + "epoch": 1.8019935015757458, + "grad_norm": 0.003040730021893978, + "learning_rate": 0.004483869030152965, + "loss": 0.1163, + "num_input_tokens_seen": 47813056, + "step": 36880 + }, + { + "epoch": 1.8022378032394402, + "grad_norm": 0.0007588575826957822, + "learning_rate": 0.004469585488327904, + "loss": 0.081, + "num_input_tokens_seen": 47819616, + "step": 36885 + }, + { + "epoch": 1.8024821049031345, + "grad_norm": 0.0014682268956676126, + "learning_rate": 0.0044553243891216395, + "loss": 0.0852, + "num_input_tokens_seen": 47825888, + "step": 36890 + }, + { + "epoch": 1.8027264065668287, + "grad_norm": 0.0009682468371465802, + "learning_rate": 0.004441085734733363, + "loss": 0.0717, + "num_input_tokens_seen": 47832544, + "step": 36895 + }, + { + "epoch": 1.802970708230523, + "grad_norm": 0.0011357032926753163, + "learning_rate": 0.004426869527358884, + "loss": 0.0904, + "num_input_tokens_seen": 47839200, + "step": 36900 + }, + { + "epoch": 1.8032150098942172, + "grad_norm": 0.0010906190145760775, + "learning_rate": 0.0044126757691905156, + "loss": 0.0953, + "num_input_tokens_seen": 47845664, + "step": 36905 + }, + { + "epoch": 1.8034593115579116, + "grad_norm": 0.001432662713341415, + "learning_rate": 0.004398504462417107, + "loss": 0.0662, + "num_input_tokens_seen": 47852000, + "step": 36910 + }, + { + "epoch": 1.803703613221606, + "grad_norm": 0.0014820838114246726, + "learning_rate": 0.0043843556092240605, + "loss": 0.0766, + "num_input_tokens_seen": 47858272, + "step": 36915 + }, + { + "epoch": 1.8039479148853004, + "grad_norm": 0.001142985885962844, + "learning_rate": 0.004370229211793281, + "loss": 0.0608, + "num_input_tokens_seen": 47864576, + "step": 36920 + }, + { + "epoch": 1.8041922165489948, + "grad_norm": 0.0010109510039910674, + "learning_rate": 0.0043561252723032405, + "loss": 0.0882, + "num_input_tokens_seen": 47870720, + "step": 36925 + }, + { + "epoch": 1.8044365182126891, + "grad_norm": 0.0014687908114865422, + "learning_rate": 0.004342043792929001, + "loss": 0.1032, + "num_input_tokens_seen": 47876800, + "step": 36930 + }, + { + "epoch": 1.8046808198763835, + "grad_norm": 0.0023686103522777557, + "learning_rate": 0.004327984775842025, + "loss": 0.0822, + "num_input_tokens_seen": 47882976, + "step": 36935 + }, + { + "epoch": 1.8049251215400777, + "grad_norm": 0.0015600724145770073, + "learning_rate": 0.004313948223210428, + "loss": 0.0838, + "num_input_tokens_seen": 47889440, + "step": 36940 + }, + { + "epoch": 1.805169423203772, + "grad_norm": 0.0014266434591263533, + "learning_rate": 0.004299934137198846, + "loss": 0.0668, + "num_input_tokens_seen": 47895584, + "step": 36945 + }, + { + "epoch": 1.8054137248674662, + "grad_norm": 0.0012375949881970882, + "learning_rate": 0.004285942519968383, + "loss": 0.083, + "num_input_tokens_seen": 47902240, + "step": 36950 + }, + { + "epoch": 1.8056580265311606, + "grad_norm": 0.0008491519256494939, + "learning_rate": 0.004271973373676746, + "loss": 0.0877, + "num_input_tokens_seen": 47908544, + "step": 36955 + }, + { + "epoch": 1.805902328194855, + "grad_norm": 0.0011123595759272575, + "learning_rate": 0.004258026700478146, + "loss": 0.1057, + "num_input_tokens_seen": 47914816, + "step": 36960 + }, + { + "epoch": 1.8061466298585493, + "grad_norm": 0.0014600037829950452, + "learning_rate": 0.004244102502523328, + "loss": 0.0837, + "num_input_tokens_seen": 47921248, + "step": 36965 + }, + { + "epoch": 1.8063909315222437, + "grad_norm": 0.0015467163175344467, + "learning_rate": 0.004230200781959592, + "loss": 0.0996, + "num_input_tokens_seen": 47927104, + "step": 36970 + }, + { + "epoch": 1.806635233185938, + "grad_norm": 0.0014562854776158929, + "learning_rate": 0.004216321540930756, + "loss": 0.074, + "num_input_tokens_seen": 47933696, + "step": 36975 + }, + { + "epoch": 1.8068795348496325, + "grad_norm": 0.0011827538255602121, + "learning_rate": 0.004202464781577175, + "loss": 0.0743, + "num_input_tokens_seen": 47940224, + "step": 36980 + }, + { + "epoch": 1.8071238365133266, + "grad_norm": 0.0018735200865194201, + "learning_rate": 0.00418863050603574, + "loss": 0.1288, + "num_input_tokens_seen": 47946272, + "step": 36985 + }, + { + "epoch": 1.807368138177021, + "grad_norm": 0.0010980179067701101, + "learning_rate": 0.004174818716439843, + "loss": 0.0781, + "num_input_tokens_seen": 47952896, + "step": 36990 + }, + { + "epoch": 1.8076124398407152, + "grad_norm": 0.0017246123170480132, + "learning_rate": 0.004161029414919464, + "loss": 0.0816, + "num_input_tokens_seen": 47959840, + "step": 36995 + }, + { + "epoch": 1.8078567415044096, + "grad_norm": 0.0009341640397906303, + "learning_rate": 0.004147262603601071, + "loss": 0.0703, + "num_input_tokens_seen": 47966176, + "step": 37000 + }, + { + "epoch": 1.8078567415044096, + "eval_loss": 0.08418054133653641, + "eval_runtime": 402.7014, + "eval_samples_per_second": 90.352, + "eval_steps_per_second": 22.59, + "num_input_tokens_seen": 47966176, + "step": 37000 + }, + { + "epoch": 1.808101043168104, + "grad_norm": 0.0014173737727105618, + "learning_rate": 0.004133518284607679, + "loss": 0.071, + "num_input_tokens_seen": 47972704, + "step": 37005 + }, + { + "epoch": 1.8083453448317983, + "grad_norm": 0.0023533354979008436, + "learning_rate": 0.004119796460058861, + "loss": 0.0864, + "num_input_tokens_seen": 47979104, + "step": 37010 + }, + { + "epoch": 1.8085896464954927, + "grad_norm": 0.00132992141880095, + "learning_rate": 0.00410609713207064, + "loss": 0.0598, + "num_input_tokens_seen": 47985760, + "step": 37015 + }, + { + "epoch": 1.808833948159187, + "grad_norm": 0.0006186366081237793, + "learning_rate": 0.004092420302755678, + "loss": 0.0685, + "num_input_tokens_seen": 47992832, + "step": 37020 + }, + { + "epoch": 1.8090782498228815, + "grad_norm": 0.0010775492992252111, + "learning_rate": 0.004078765974223103, + "loss": 0.0525, + "num_input_tokens_seen": 47998720, + "step": 37025 + }, + { + "epoch": 1.8093225514865756, + "grad_norm": 0.0010820245370268822, + "learning_rate": 0.004065134148578564, + "loss": 0.0892, + "num_input_tokens_seen": 48004864, + "step": 37030 + }, + { + "epoch": 1.80956685315027, + "grad_norm": 0.0010411362163722515, + "learning_rate": 0.004051524827924279, + "loss": 0.0703, + "num_input_tokens_seen": 48011424, + "step": 37035 + }, + { + "epoch": 1.8098111548139642, + "grad_norm": 0.0015403198776766658, + "learning_rate": 0.004037938014358955, + "loss": 0.0812, + "num_input_tokens_seen": 48017696, + "step": 37040 + }, + { + "epoch": 1.8100554564776585, + "grad_norm": 0.0010873832507058978, + "learning_rate": 0.004024373709977863, + "loss": 0.0863, + "num_input_tokens_seen": 48024512, + "step": 37045 + }, + { + "epoch": 1.810299758141353, + "grad_norm": 0.0014822253724560142, + "learning_rate": 0.004010831916872814, + "loss": 0.0791, + "num_input_tokens_seen": 48031392, + "step": 37050 + }, + { + "epoch": 1.8105440598050473, + "grad_norm": 0.001120224827900529, + "learning_rate": 0.003997312637132089, + "loss": 0.0688, + "num_input_tokens_seen": 48037888, + "step": 37055 + }, + { + "epoch": 1.8107883614687417, + "grad_norm": 0.001441140309907496, + "learning_rate": 0.003983815872840535, + "loss": 0.1148, + "num_input_tokens_seen": 48044160, + "step": 37060 + }, + { + "epoch": 1.811032663132436, + "grad_norm": 0.001679253880865872, + "learning_rate": 0.003970341626079521, + "loss": 0.09, + "num_input_tokens_seen": 48050592, + "step": 37065 + }, + { + "epoch": 1.8112769647961302, + "grad_norm": 0.0015953710535541177, + "learning_rate": 0.003956889898926952, + "loss": 0.0982, + "num_input_tokens_seen": 48056800, + "step": 37070 + }, + { + "epoch": 1.8115212664598246, + "grad_norm": 0.0020031528547406197, + "learning_rate": 0.0039434606934572675, + "loss": 0.086, + "num_input_tokens_seen": 48063424, + "step": 37075 + }, + { + "epoch": 1.811765568123519, + "grad_norm": 0.0009619435877539217, + "learning_rate": 0.003930054011741396, + "loss": 0.0851, + "num_input_tokens_seen": 48069984, + "step": 37080 + }, + { + "epoch": 1.8120098697872131, + "grad_norm": 0.0017621676670387387, + "learning_rate": 0.0039166698558468155, + "loss": 0.0695, + "num_input_tokens_seen": 48076608, + "step": 37085 + }, + { + "epoch": 1.8122541714509075, + "grad_norm": 0.0015572239644825459, + "learning_rate": 0.0039033082278375594, + "loss": 0.0672, + "num_input_tokens_seen": 48083360, + "step": 37090 + }, + { + "epoch": 1.8124984731146019, + "grad_norm": 0.0012503809994086623, + "learning_rate": 0.003889969129774112, + "loss": 0.0756, + "num_input_tokens_seen": 48089440, + "step": 37095 + }, + { + "epoch": 1.8127427747782963, + "grad_norm": 0.000468697544420138, + "learning_rate": 0.0038766525637135784, + "loss": 0.0875, + "num_input_tokens_seen": 48096000, + "step": 37100 + }, + { + "epoch": 1.8129870764419906, + "grad_norm": 0.0007359639857895672, + "learning_rate": 0.0038633585317095318, + "loss": 0.0777, + "num_input_tokens_seen": 48102752, + "step": 37105 + }, + { + "epoch": 1.813231378105685, + "grad_norm": 0.001051852130331099, + "learning_rate": 0.00385008703581205, + "loss": 0.0702, + "num_input_tokens_seen": 48109152, + "step": 37110 + }, + { + "epoch": 1.8134756797693792, + "grad_norm": 0.0011142693692818284, + "learning_rate": 0.0038368380780677944, + "loss": 0.1126, + "num_input_tokens_seen": 48115360, + "step": 37115 + }, + { + "epoch": 1.8137199814330736, + "grad_norm": 0.0009514515986666083, + "learning_rate": 0.003823611660519882, + "loss": 0.056, + "num_input_tokens_seen": 48121856, + "step": 37120 + }, + { + "epoch": 1.813964283096768, + "grad_norm": 0.0011967828031629324, + "learning_rate": 0.0038104077852080475, + "loss": 0.0758, + "num_input_tokens_seen": 48128064, + "step": 37125 + }, + { + "epoch": 1.814208584760462, + "grad_norm": 0.0010423845378682017, + "learning_rate": 0.003797226454168462, + "loss": 0.0716, + "num_input_tokens_seen": 48134752, + "step": 37130 + }, + { + "epoch": 1.8144528864241565, + "grad_norm": 0.0011526737362146378, + "learning_rate": 0.003784067669433849, + "loss": 0.0814, + "num_input_tokens_seen": 48141472, + "step": 37135 + }, + { + "epoch": 1.8146971880878509, + "grad_norm": 0.0019530962454155087, + "learning_rate": 0.0037709314330334528, + "loss": 0.0873, + "num_input_tokens_seen": 48147808, + "step": 37140 + }, + { + "epoch": 1.8149414897515452, + "grad_norm": 0.001136021688580513, + "learning_rate": 0.003757817746993086, + "loss": 0.077, + "num_input_tokens_seen": 48154432, + "step": 37145 + }, + { + "epoch": 1.8151857914152396, + "grad_norm": 0.0018289118306711316, + "learning_rate": 0.0037447266133349977, + "loss": 0.0905, + "num_input_tokens_seen": 48160736, + "step": 37150 + }, + { + "epoch": 1.815430093078934, + "grad_norm": 0.002001103013753891, + "learning_rate": 0.003731658034078039, + "loss": 0.0871, + "num_input_tokens_seen": 48167392, + "step": 37155 + }, + { + "epoch": 1.8156743947426282, + "grad_norm": 0.0007952868472784758, + "learning_rate": 0.0037186120112375153, + "loss": 0.0743, + "num_input_tokens_seen": 48174048, + "step": 37160 + }, + { + "epoch": 1.8159186964063225, + "grad_norm": 0.0015973318368196487, + "learning_rate": 0.003705588546825317, + "loss": 0.0674, + "num_input_tokens_seen": 48180928, + "step": 37165 + }, + { + "epoch": 1.816162998070017, + "grad_norm": 0.0012942341854795814, + "learning_rate": 0.0036925876428498205, + "loss": 0.0647, + "num_input_tokens_seen": 48187264, + "step": 37170 + }, + { + "epoch": 1.816407299733711, + "grad_norm": 0.0009273190516978502, + "learning_rate": 0.0036796093013159057, + "loss": 0.0676, + "num_input_tokens_seen": 48193408, + "step": 37175 + }, + { + "epoch": 1.8166516013974054, + "grad_norm": 0.0020604629535228014, + "learning_rate": 0.0036666535242250217, + "loss": 0.0844, + "num_input_tokens_seen": 48200128, + "step": 37180 + }, + { + "epoch": 1.8168959030610998, + "grad_norm": 0.000795457512140274, + "learning_rate": 0.003653720313575104, + "loss": 0.0948, + "num_input_tokens_seen": 48206816, + "step": 37185 + }, + { + "epoch": 1.8171402047247942, + "grad_norm": 0.0013278432888910174, + "learning_rate": 0.003640809671360623, + "loss": 0.0834, + "num_input_tokens_seen": 48213312, + "step": 37190 + }, + { + "epoch": 1.8173845063884886, + "grad_norm": 0.001866507576778531, + "learning_rate": 0.003627921599572553, + "loss": 0.0854, + "num_input_tokens_seen": 48220800, + "step": 37195 + }, + { + "epoch": 1.817628808052183, + "grad_norm": 0.0010675728553906083, + "learning_rate": 0.003615056100198405, + "loss": 0.0636, + "num_input_tokens_seen": 48227328, + "step": 37200 + }, + { + "epoch": 1.817628808052183, + "eval_loss": 0.08353738486766815, + "eval_runtime": 402.1971, + "eval_samples_per_second": 90.466, + "eval_steps_per_second": 22.618, + "num_input_tokens_seen": 48227328, + "step": 37200 + }, + { + "epoch": 1.8178731097158771, + "grad_norm": 0.0010341460583731532, + "learning_rate": 0.003602213175222174, + "loss": 0.0612, + "num_input_tokens_seen": 48233984, + "step": 37205 + }, + { + "epoch": 1.8181174113795715, + "grad_norm": 0.0016262144781649113, + "learning_rate": 0.0035893928266244432, + "loss": 0.0839, + "num_input_tokens_seen": 48240416, + "step": 37210 + }, + { + "epoch": 1.8183617130432657, + "grad_norm": 0.001545254490338266, + "learning_rate": 0.003576595056382248, + "loss": 0.1144, + "num_input_tokens_seen": 48246624, + "step": 37215 + }, + { + "epoch": 1.81860601470696, + "grad_norm": 0.0007502130465582013, + "learning_rate": 0.0035638198664691423, + "loss": 0.0586, + "num_input_tokens_seen": 48253824, + "step": 37220 + }, + { + "epoch": 1.8188503163706544, + "grad_norm": 0.0017530926270410419, + "learning_rate": 0.003551067258855267, + "loss": 0.0881, + "num_input_tokens_seen": 48260256, + "step": 37225 + }, + { + "epoch": 1.8190946180343488, + "grad_norm": 0.0018382598645985126, + "learning_rate": 0.0035383372355071996, + "loss": 0.0929, + "num_input_tokens_seen": 48266912, + "step": 37230 + }, + { + "epoch": 1.8193389196980432, + "grad_norm": 0.001145727001130581, + "learning_rate": 0.0035256297983881023, + "loss": 0.0675, + "num_input_tokens_seen": 48273024, + "step": 37235 + }, + { + "epoch": 1.8195832213617376, + "grad_norm": 0.001369217410683632, + "learning_rate": 0.0035129449494575747, + "loss": 0.0936, + "num_input_tokens_seen": 48279584, + "step": 37240 + }, + { + "epoch": 1.819827523025432, + "grad_norm": 0.0018467794870957732, + "learning_rate": 0.0035002826906718187, + "loss": 0.0898, + "num_input_tokens_seen": 48285632, + "step": 37245 + }, + { + "epoch": 1.820071824689126, + "grad_norm": 0.0011584404855966568, + "learning_rate": 0.003487643023983522, + "loss": 0.0712, + "num_input_tokens_seen": 48292096, + "step": 37250 + }, + { + "epoch": 1.8203161263528205, + "grad_norm": 0.0011987106408923864, + "learning_rate": 0.003475025951341842, + "loss": 0.0788, + "num_input_tokens_seen": 48298528, + "step": 37255 + }, + { + "epoch": 1.8205604280165146, + "grad_norm": 0.001295343623496592, + "learning_rate": 0.00346243147469249, + "loss": 0.0747, + "num_input_tokens_seen": 48304928, + "step": 37260 + }, + { + "epoch": 1.820804729680209, + "grad_norm": 0.0011851555900648236, + "learning_rate": 0.0034498595959777446, + "loss": 0.1027, + "num_input_tokens_seen": 48311424, + "step": 37265 + }, + { + "epoch": 1.8210490313439034, + "grad_norm": 0.0013249441981315613, + "learning_rate": 0.003437310317136305, + "loss": 0.0767, + "num_input_tokens_seen": 48317792, + "step": 37270 + }, + { + "epoch": 1.8212933330075978, + "grad_norm": 0.0011302989441901445, + "learning_rate": 0.0034247836401034236, + "loss": 0.099, + "num_input_tokens_seen": 48324160, + "step": 37275 + }, + { + "epoch": 1.8215376346712921, + "grad_norm": 0.0010580412345007062, + "learning_rate": 0.003412279566810905, + "loss": 0.0569, + "num_input_tokens_seen": 48330624, + "step": 37280 + }, + { + "epoch": 1.8217819363349865, + "grad_norm": 0.0010574000189080834, + "learning_rate": 0.00339979809918699, + "loss": 0.0972, + "num_input_tokens_seen": 48337376, + "step": 37285 + }, + { + "epoch": 1.822026237998681, + "grad_norm": 0.0011168959317728877, + "learning_rate": 0.0033873392391565228, + "loss": 0.0712, + "num_input_tokens_seen": 48343424, + "step": 37290 + }, + { + "epoch": 1.822270539662375, + "grad_norm": 0.0013071574503555894, + "learning_rate": 0.003374902988640782, + "loss": 0.0722, + "num_input_tokens_seen": 48350112, + "step": 37295 + }, + { + "epoch": 1.8225148413260694, + "grad_norm": 0.0016852106200531125, + "learning_rate": 0.0033624893495576014, + "loss": 0.0945, + "num_input_tokens_seen": 48356416, + "step": 37300 + }, + { + "epoch": 1.8227591429897636, + "grad_norm": 0.001959822839125991, + "learning_rate": 0.0033500983238213323, + "loss": 0.1069, + "num_input_tokens_seen": 48363136, + "step": 37305 + }, + { + "epoch": 1.823003444653458, + "grad_norm": 0.0013464662479236722, + "learning_rate": 0.0033377299133428126, + "loss": 0.0732, + "num_input_tokens_seen": 48369696, + "step": 37310 + }, + { + "epoch": 1.8232477463171524, + "grad_norm": 0.0010192171903327107, + "learning_rate": 0.003325384120029434, + "loss": 0.0852, + "num_input_tokens_seen": 48376320, + "step": 37315 + }, + { + "epoch": 1.8234920479808467, + "grad_norm": 0.0009927006904035807, + "learning_rate": 0.0033130609457850233, + "loss": 0.0849, + "num_input_tokens_seen": 48382496, + "step": 37320 + }, + { + "epoch": 1.8237363496445411, + "grad_norm": 0.0016513204900547862, + "learning_rate": 0.0033007603925100104, + "loss": 0.0744, + "num_input_tokens_seen": 48389504, + "step": 37325 + }, + { + "epoch": 1.8239806513082355, + "grad_norm": 0.001636513159610331, + "learning_rate": 0.003288482462101294, + "loss": 0.083, + "num_input_tokens_seen": 48396000, + "step": 37330 + }, + { + "epoch": 1.8242249529719299, + "grad_norm": 0.0017273086123168468, + "learning_rate": 0.0032762271564522605, + "loss": 0.0595, + "num_input_tokens_seen": 48402496, + "step": 37335 + }, + { + "epoch": 1.824469254635624, + "grad_norm": 0.0018288959981873631, + "learning_rate": 0.003263994477452864, + "loss": 0.0898, + "num_input_tokens_seen": 48409344, + "step": 37340 + }, + { + "epoch": 1.8247135562993184, + "grad_norm": 0.0020622711163014174, + "learning_rate": 0.0032517844269895125, + "loss": 0.0789, + "num_input_tokens_seen": 48416384, + "step": 37345 + }, + { + "epoch": 1.8249578579630126, + "grad_norm": 0.002552326302975416, + "learning_rate": 0.0032395970069451496, + "loss": 0.1115, + "num_input_tokens_seen": 48423008, + "step": 37350 + }, + { + "epoch": 1.825202159626707, + "grad_norm": 0.0024663235526531935, + "learning_rate": 0.0032274322191992388, + "loss": 0.0876, + "num_input_tokens_seen": 48429056, + "step": 37355 + }, + { + "epoch": 1.8254464612904013, + "grad_norm": 0.0015793760539963841, + "learning_rate": 0.0032152900656277294, + "loss": 0.0734, + "num_input_tokens_seen": 48435392, + "step": 37360 + }, + { + "epoch": 1.8256907629540957, + "grad_norm": 0.0009412110666744411, + "learning_rate": 0.0032031705481030902, + "loss": 0.0762, + "num_input_tokens_seen": 48441536, + "step": 37365 + }, + { + "epoch": 1.82593506461779, + "grad_norm": 0.001691531972028315, + "learning_rate": 0.0031910736684943428, + "loss": 0.1049, + "num_input_tokens_seen": 48448064, + "step": 37370 + }, + { + "epoch": 1.8261793662814845, + "grad_norm": 0.0008612447418272495, + "learning_rate": 0.0031789994286669453, + "loss": 0.0874, + "num_input_tokens_seen": 48453920, + "step": 37375 + }, + { + "epoch": 1.8264236679451789, + "grad_norm": 0.001393828890286386, + "learning_rate": 0.003166947830482908, + "loss": 0.0833, + "num_input_tokens_seen": 48460000, + "step": 37380 + }, + { + "epoch": 1.826667969608873, + "grad_norm": 0.0018660466885194182, + "learning_rate": 0.003154918875800727, + "loss": 0.0912, + "num_input_tokens_seen": 48466784, + "step": 37385 + }, + { + "epoch": 1.8269122712725674, + "grad_norm": 0.0013537879567593336, + "learning_rate": 0.00314291256647542, + "loss": 0.1008, + "num_input_tokens_seen": 48472800, + "step": 37390 + }, + { + "epoch": 1.8271565729362615, + "grad_norm": 0.0019149596337229013, + "learning_rate": 0.0031309289043585375, + "loss": 0.0924, + "num_input_tokens_seen": 48479040, + "step": 37395 + }, + { + "epoch": 1.827400874599956, + "grad_norm": 0.0017431918531656265, + "learning_rate": 0.003118967891298069, + "loss": 0.0774, + "num_input_tokens_seen": 48485632, + "step": 37400 + }, + { + "epoch": 1.827400874599956, + "eval_loss": 0.08353451639413834, + "eval_runtime": 402.301, + "eval_samples_per_second": 90.442, + "eval_steps_per_second": 22.612, + "num_input_tokens_seen": 48485632, + "step": 37400 + }, + { + "epoch": 1.8276451762636503, + "grad_norm": 0.0018287822604179382, + "learning_rate": 0.003107029529138572, + "loss": 0.0807, + "num_input_tokens_seen": 48492032, + "step": 37405 + }, + { + "epoch": 1.8278894779273447, + "grad_norm": 0.0008201120654121041, + "learning_rate": 0.0030951138197211235, + "loss": 0.087, + "num_input_tokens_seen": 48498752, + "step": 37410 + }, + { + "epoch": 1.828133779591039, + "grad_norm": 0.0012955744750797749, + "learning_rate": 0.0030832207648832377, + "loss": 0.092, + "num_input_tokens_seen": 48505312, + "step": 37415 + }, + { + "epoch": 1.8283780812547334, + "grad_norm": 0.0007369329687207937, + "learning_rate": 0.0030713503664589635, + "loss": 0.0815, + "num_input_tokens_seen": 48511776, + "step": 37420 + }, + { + "epoch": 1.8286223829184278, + "grad_norm": 0.0017562313005328178, + "learning_rate": 0.0030595026262788872, + "loss": 0.0777, + "num_input_tokens_seen": 48518272, + "step": 37425 + }, + { + "epoch": 1.828866684582122, + "grad_norm": 0.0019045755034312606, + "learning_rate": 0.00304767754617008, + "loss": 0.0901, + "num_input_tokens_seen": 48524448, + "step": 37430 + }, + { + "epoch": 1.8291109862458164, + "grad_norm": 0.002598862862214446, + "learning_rate": 0.003035875127956117, + "loss": 0.0969, + "num_input_tokens_seen": 48530560, + "step": 37435 + }, + { + "epoch": 1.8293552879095105, + "grad_norm": 0.0012649450218304992, + "learning_rate": 0.0030240953734570752, + "loss": 0.0881, + "num_input_tokens_seen": 48537376, + "step": 37440 + }, + { + "epoch": 1.829599589573205, + "grad_norm": 0.0011773357400670648, + "learning_rate": 0.003012338284489535, + "loss": 0.0792, + "num_input_tokens_seen": 48543904, + "step": 37445 + }, + { + "epoch": 1.8298438912368993, + "grad_norm": 0.00109123345464468, + "learning_rate": 0.0030006038628665964, + "loss": 0.0796, + "num_input_tokens_seen": 48550848, + "step": 37450 + }, + { + "epoch": 1.8300881929005937, + "grad_norm": 0.0012116336729377508, + "learning_rate": 0.002988892110397845, + "loss": 0.079, + "num_input_tokens_seen": 48557024, + "step": 37455 + }, + { + "epoch": 1.830332494564288, + "grad_norm": 0.0017197419656440616, + "learning_rate": 0.0029772030288894025, + "loss": 0.0979, + "num_input_tokens_seen": 48563136, + "step": 37460 + }, + { + "epoch": 1.8305767962279824, + "grad_norm": 0.0008352402946911752, + "learning_rate": 0.0029655366201438438, + "loss": 0.0647, + "num_input_tokens_seen": 48569824, + "step": 37465 + }, + { + "epoch": 1.8308210978916768, + "grad_norm": 0.0015222460497170687, + "learning_rate": 0.0029538928859602965, + "loss": 0.0909, + "num_input_tokens_seen": 48575968, + "step": 37470 + }, + { + "epoch": 1.831065399555371, + "grad_norm": 0.0012349736643955112, + "learning_rate": 0.002942271828134374, + "loss": 0.0804, + "num_input_tokens_seen": 48582560, + "step": 37475 + }, + { + "epoch": 1.8313097012190653, + "grad_norm": 0.0008193748653866351, + "learning_rate": 0.00293067344845816, + "loss": 0.0759, + "num_input_tokens_seen": 48589376, + "step": 37480 + }, + { + "epoch": 1.8315540028827595, + "grad_norm": 0.0008042271947488189, + "learning_rate": 0.0029190977487202896, + "loss": 0.0504, + "num_input_tokens_seen": 48595712, + "step": 37485 + }, + { + "epoch": 1.8317983045464539, + "grad_norm": 0.001165915746241808, + "learning_rate": 0.0029075447307058853, + "loss": 0.0864, + "num_input_tokens_seen": 48602176, + "step": 37490 + }, + { + "epoch": 1.8320426062101483, + "grad_norm": 0.0011153790401294827, + "learning_rate": 0.0028960143961965722, + "loss": 0.0733, + "num_input_tokens_seen": 48608640, + "step": 37495 + }, + { + "epoch": 1.8322869078738426, + "grad_norm": 0.001260282821021974, + "learning_rate": 0.002884506746970461, + "loss": 0.0954, + "num_input_tokens_seen": 48614944, + "step": 37500 + }, + { + "epoch": 1.832531209537537, + "grad_norm": 0.001829613815061748, + "learning_rate": 0.0028730217848021654, + "loss": 0.0713, + "num_input_tokens_seen": 48621664, + "step": 37505 + }, + { + "epoch": 1.8327755112012314, + "grad_norm": 0.0013398312730714679, + "learning_rate": 0.0028615595114628188, + "loss": 0.0837, + "num_input_tokens_seen": 48628192, + "step": 37510 + }, + { + "epoch": 1.8330198128649258, + "grad_norm": 0.0014067365555092692, + "learning_rate": 0.002850119928720074, + "loss": 0.0689, + "num_input_tokens_seen": 48634368, + "step": 37515 + }, + { + "epoch": 1.83326411452862, + "grad_norm": 0.0011682199547067285, + "learning_rate": 0.0028387030383380195, + "loss": 0.0791, + "num_input_tokens_seen": 48641248, + "step": 37520 + }, + { + "epoch": 1.8335084161923143, + "grad_norm": 0.0008021810790523887, + "learning_rate": 0.0028273088420772974, + "loss": 0.0708, + "num_input_tokens_seen": 48647680, + "step": 37525 + }, + { + "epoch": 1.8337527178560085, + "grad_norm": 0.001999689731746912, + "learning_rate": 0.002815937341695068, + "loss": 0.1022, + "num_input_tokens_seen": 48653920, + "step": 37530 + }, + { + "epoch": 1.8339970195197028, + "grad_norm": 0.0017586550675332546, + "learning_rate": 0.0028045885389448963, + "loss": 0.0717, + "num_input_tokens_seen": 48660256, + "step": 37535 + }, + { + "epoch": 1.8342413211833972, + "grad_norm": 0.0010357069550082088, + "learning_rate": 0.002793262435576965, + "loss": 0.0823, + "num_input_tokens_seen": 48666976, + "step": 37540 + }, + { + "epoch": 1.8344856228470916, + "grad_norm": 0.0017430359730497003, + "learning_rate": 0.0027819590333378772, + "loss": 0.074, + "num_input_tokens_seen": 48673280, + "step": 37545 + }, + { + "epoch": 1.834729924510786, + "grad_norm": 0.001053321291692555, + "learning_rate": 0.002770678333970755, + "loss": 0.0675, + "num_input_tokens_seen": 48679904, + "step": 37550 + }, + { + "epoch": 1.8349742261744804, + "grad_norm": 0.0009871566435322165, + "learning_rate": 0.0027594203392152573, + "loss": 0.0725, + "num_input_tokens_seen": 48686272, + "step": 37555 + }, + { + "epoch": 1.8352185278381747, + "grad_norm": 0.0008603817550465465, + "learning_rate": 0.002748185050807478, + "loss": 0.0693, + "num_input_tokens_seen": 48692384, + "step": 37560 + }, + { + "epoch": 1.835462829501869, + "grad_norm": 0.0016051350394263864, + "learning_rate": 0.002736972470480031, + "loss": 0.0909, + "num_input_tokens_seen": 48699104, + "step": 37565 + }, + { + "epoch": 1.8357071311655633, + "grad_norm": 0.0016163127729669213, + "learning_rate": 0.002725782599962068, + "loss": 0.0571, + "num_input_tokens_seen": 48706016, + "step": 37570 + }, + { + "epoch": 1.8359514328292574, + "grad_norm": 0.0006502077449113131, + "learning_rate": 0.0027146154409791734, + "loss": 0.0588, + "num_input_tokens_seen": 48712992, + "step": 37575 + }, + { + "epoch": 1.8361957344929518, + "grad_norm": 0.0011676240246742964, + "learning_rate": 0.002703470995253504, + "loss": 0.0696, + "num_input_tokens_seen": 48719072, + "step": 37580 + }, + { + "epoch": 1.8364400361566462, + "grad_norm": 0.0008516762172803283, + "learning_rate": 0.0026923492645036184, + "loss": 0.0832, + "num_input_tokens_seen": 48725632, + "step": 37585 + }, + { + "epoch": 1.8366843378203406, + "grad_norm": 0.0010326928459107876, + "learning_rate": 0.0026812502504446776, + "loss": 0.065, + "num_input_tokens_seen": 48732000, + "step": 37590 + }, + { + "epoch": 1.836928639484035, + "grad_norm": 0.0016208275919780135, + "learning_rate": 0.0026701739547882798, + "loss": 0.1016, + "num_input_tokens_seen": 48738272, + "step": 37595 + }, + { + "epoch": 1.8371729411477293, + "grad_norm": 0.0007013199501670897, + "learning_rate": 0.0026591203792425077, + "loss": 0.0662, + "num_input_tokens_seen": 48744768, + "step": 37600 + }, + { + "epoch": 1.8371729411477293, + "eval_loss": 0.0835486650466919, + "eval_runtime": 402.6305, + "eval_samples_per_second": 90.368, + "eval_steps_per_second": 22.594, + "num_input_tokens_seen": 48744768, + "step": 37600 + }, + { + "epoch": 1.8374172428114237, + "grad_norm": 0.0036647762171924114, + "learning_rate": 0.0026480895255119818, + "loss": 0.0896, + "num_input_tokens_seen": 48750912, + "step": 37605 + }, + { + "epoch": 1.8376615444751179, + "grad_norm": 0.0012667940463870764, + "learning_rate": 0.002637081395297791, + "loss": 0.0761, + "num_input_tokens_seen": 48757472, + "step": 37610 + }, + { + "epoch": 1.8379058461388122, + "grad_norm": 0.0008520680130459368, + "learning_rate": 0.0026260959902975113, + "loss": 0.068, + "num_input_tokens_seen": 48764000, + "step": 37615 + }, + { + "epoch": 1.8381501478025064, + "grad_norm": 0.0008208427461795509, + "learning_rate": 0.00261513331220527, + "loss": 0.0766, + "num_input_tokens_seen": 48770784, + "step": 37620 + }, + { + "epoch": 1.8383944494662008, + "grad_norm": 0.001128622330725193, + "learning_rate": 0.0026041933627116154, + "loss": 0.0928, + "num_input_tokens_seen": 48776960, + "step": 37625 + }, + { + "epoch": 1.8386387511298952, + "grad_norm": 0.001748298411257565, + "learning_rate": 0.0025932761435036476, + "loss": 0.0462, + "num_input_tokens_seen": 48783776, + "step": 37630 + }, + { + "epoch": 1.8388830527935895, + "grad_norm": 0.00149906100705266, + "learning_rate": 0.002582381656264904, + "loss": 0.0711, + "num_input_tokens_seen": 48790112, + "step": 37635 + }, + { + "epoch": 1.839127354457284, + "grad_norm": 0.0021265242248773575, + "learning_rate": 0.0025715099026754895, + "loss": 0.116, + "num_input_tokens_seen": 48796512, + "step": 37640 + }, + { + "epoch": 1.8393716561209783, + "grad_norm": 0.0024541516322642565, + "learning_rate": 0.002560660884411947, + "loss": 0.1155, + "num_input_tokens_seen": 48802688, + "step": 37645 + }, + { + "epoch": 1.8396159577846725, + "grad_norm": 0.0016540053766220808, + "learning_rate": 0.0025498346031473385, + "loss": 0.0971, + "num_input_tokens_seen": 48809600, + "step": 37650 + }, + { + "epoch": 1.8398602594483668, + "grad_norm": 0.0013929131673648953, + "learning_rate": 0.0025390310605511945, + "loss": 0.0753, + "num_input_tokens_seen": 48815904, + "step": 37655 + }, + { + "epoch": 1.8401045611120612, + "grad_norm": 0.0016911699203774333, + "learning_rate": 0.0025282502582895995, + "loss": 0.1177, + "num_input_tokens_seen": 48822272, + "step": 37660 + }, + { + "epoch": 1.8403488627757554, + "grad_norm": 0.001503700390458107, + "learning_rate": 0.002517492198025023, + "loss": 0.0994, + "num_input_tokens_seen": 48828832, + "step": 37665 + }, + { + "epoch": 1.8405931644394498, + "grad_norm": 0.001121738227084279, + "learning_rate": 0.0025067568814165554, + "loss": 0.0623, + "num_input_tokens_seen": 48835200, + "step": 37670 + }, + { + "epoch": 1.8408374661031441, + "grad_norm": 0.0014181106816977262, + "learning_rate": 0.0024960443101196884, + "loss": 0.0878, + "num_input_tokens_seen": 48841376, + "step": 37675 + }, + { + "epoch": 1.8410817677668385, + "grad_norm": 0.0013371383538469672, + "learning_rate": 0.002485354485786434, + "loss": 0.0784, + "num_input_tokens_seen": 48847584, + "step": 37680 + }, + { + "epoch": 1.841326069430533, + "grad_norm": 0.0013555475743487477, + "learning_rate": 0.002474687410065307, + "loss": 0.0681, + "num_input_tokens_seen": 48854048, + "step": 37685 + }, + { + "epoch": 1.8415703710942273, + "grad_norm": 0.0012967863585799932, + "learning_rate": 0.002464043084601308, + "loss": 0.1003, + "num_input_tokens_seen": 48860672, + "step": 37690 + }, + { + "epoch": 1.8418146727579214, + "grad_norm": 0.0011629087384790182, + "learning_rate": 0.0024534215110358915, + "loss": 0.0737, + "num_input_tokens_seen": 48867232, + "step": 37695 + }, + { + "epoch": 1.8420589744216158, + "grad_norm": 0.0018191009294241667, + "learning_rate": 0.002442822691007096, + "loss": 0.0733, + "num_input_tokens_seen": 48873536, + "step": 37700 + }, + { + "epoch": 1.8423032760853102, + "grad_norm": 0.0012553707929328084, + "learning_rate": 0.002432246626149348, + "loss": 0.0894, + "num_input_tokens_seen": 48880320, + "step": 37705 + }, + { + "epoch": 1.8425475777490044, + "grad_norm": 0.001010271837003529, + "learning_rate": 0.002421693318093626, + "loss": 0.0877, + "num_input_tokens_seen": 48886816, + "step": 37710 + }, + { + "epoch": 1.8427918794126987, + "grad_norm": 0.0011241771280765533, + "learning_rate": 0.0024111627684673784, + "loss": 0.0866, + "num_input_tokens_seen": 48892960, + "step": 37715 + }, + { + "epoch": 1.843036181076393, + "grad_norm": 0.0015366589650511742, + "learning_rate": 0.0024006549788945395, + "loss": 0.0958, + "num_input_tokens_seen": 48899712, + "step": 37720 + }, + { + "epoch": 1.8432804827400875, + "grad_norm": 0.0015209862031042576, + "learning_rate": 0.0023901699509955463, + "loss": 0.058, + "num_input_tokens_seen": 48906048, + "step": 37725 + }, + { + "epoch": 1.8435247844037819, + "grad_norm": 0.0018869804916903377, + "learning_rate": 0.0023797076863873554, + "loss": 0.0774, + "num_input_tokens_seen": 48912160, + "step": 37730 + }, + { + "epoch": 1.8437690860674762, + "grad_norm": 0.0016802677419036627, + "learning_rate": 0.0023692681866833262, + "loss": 0.0774, + "num_input_tokens_seen": 48918656, + "step": 37735 + }, + { + "epoch": 1.8440133877311704, + "grad_norm": 0.0007530960720032454, + "learning_rate": 0.0023588514534934046, + "loss": 0.0528, + "num_input_tokens_seen": 48925120, + "step": 37740 + }, + { + "epoch": 1.8442576893948648, + "grad_norm": 0.0012719091027975082, + "learning_rate": 0.002348457488423955, + "loss": 0.0703, + "num_input_tokens_seen": 48931392, + "step": 37745 + }, + { + "epoch": 1.844501991058559, + "grad_norm": 0.0012737780343741179, + "learning_rate": 0.0023380862930778624, + "loss": 0.1104, + "num_input_tokens_seen": 48937504, + "step": 37750 + }, + { + "epoch": 1.8447462927222533, + "grad_norm": 0.0018825207371264696, + "learning_rate": 0.0023277378690545135, + "loss": 0.0756, + "num_input_tokens_seen": 48944160, + "step": 37755 + }, + { + "epoch": 1.8449905943859477, + "grad_norm": 0.0013929478591307998, + "learning_rate": 0.0023174122179497325, + "loss": 0.0855, + "num_input_tokens_seen": 48950464, + "step": 37760 + }, + { + "epoch": 1.845234896049642, + "grad_norm": 0.0007625172147527337, + "learning_rate": 0.0023071093413558784, + "loss": 0.0698, + "num_input_tokens_seen": 48956832, + "step": 37765 + }, + { + "epoch": 1.8454791977133365, + "grad_norm": 0.001255088485777378, + "learning_rate": 0.002296829240861814, + "loss": 0.0779, + "num_input_tokens_seen": 48963392, + "step": 37770 + }, + { + "epoch": 1.8457234993770308, + "grad_norm": 0.0024794884957373142, + "learning_rate": 0.002286571918052821, + "loss": 0.0968, + "num_input_tokens_seen": 48969856, + "step": 37775 + }, + { + "epoch": 1.8459678010407252, + "grad_norm": 0.0019410214154049754, + "learning_rate": 0.0022763373745107174, + "loss": 0.1048, + "num_input_tokens_seen": 48975936, + "step": 37780 + }, + { + "epoch": 1.8462121027044194, + "grad_norm": 0.0021670556161552668, + "learning_rate": 0.0022661256118138074, + "loss": 0.103, + "num_input_tokens_seen": 48982496, + "step": 37785 + }, + { + "epoch": 1.8464564043681138, + "grad_norm": 0.000876557664014399, + "learning_rate": 0.0022559366315368645, + "loss": 0.0542, + "num_input_tokens_seen": 48988960, + "step": 37790 + }, + { + "epoch": 1.846700706031808, + "grad_norm": 0.0013284111628308892, + "learning_rate": 0.002245770435251182, + "loss": 0.0731, + "num_input_tokens_seen": 48995552, + "step": 37795 + }, + { + "epoch": 1.8469450076955023, + "grad_norm": 0.0011173420352861285, + "learning_rate": 0.002235627024524456, + "loss": 0.0792, + "num_input_tokens_seen": 49002400, + "step": 37800 + }, + { + "epoch": 1.8469450076955023, + "eval_loss": 0.0836246907711029, + "eval_runtime": 402.4898, + "eval_samples_per_second": 90.4, + "eval_steps_per_second": 22.602, + "num_input_tokens_seen": 49002400, + "step": 37800 + }, + { + "epoch": 1.8471893093591967, + "grad_norm": 0.0012058857828378677, + "learning_rate": 0.0022255064009209847, + "loss": 0.0902, + "num_input_tokens_seen": 49009024, + "step": 37805 + }, + { + "epoch": 1.847433611022891, + "grad_norm": 0.0008124314481392503, + "learning_rate": 0.0022154085660014864, + "loss": 0.0682, + "num_input_tokens_seen": 49015648, + "step": 37810 + }, + { + "epoch": 1.8476779126865854, + "grad_norm": 0.0015821902779862285, + "learning_rate": 0.0022053335213231494, + "loss": 0.0831, + "num_input_tokens_seen": 49022496, + "step": 37815 + }, + { + "epoch": 1.8479222143502798, + "grad_norm": 0.0008443401893600821, + "learning_rate": 0.002195281268439697, + "loss": 0.0634, + "num_input_tokens_seen": 49028704, + "step": 37820 + }, + { + "epoch": 1.8481665160139742, + "grad_norm": 0.00132600381039083, + "learning_rate": 0.002185251808901306, + "loss": 0.088, + "num_input_tokens_seen": 49034912, + "step": 37825 + }, + { + "epoch": 1.8484108176776683, + "grad_norm": 0.0013409185921773314, + "learning_rate": 0.0021752451442546227, + "loss": 0.0726, + "num_input_tokens_seen": 49041024, + "step": 37830 + }, + { + "epoch": 1.8486551193413627, + "grad_norm": 0.002052492694929242, + "learning_rate": 0.0021652612760428456, + "loss": 0.0971, + "num_input_tokens_seen": 49047168, + "step": 37835 + }, + { + "epoch": 1.8488994210050569, + "grad_norm": 0.0010056319879367948, + "learning_rate": 0.0021553002058055603, + "loss": 0.0617, + "num_input_tokens_seen": 49053568, + "step": 37840 + }, + { + "epoch": 1.8491437226687513, + "grad_norm": 0.0017705457285046577, + "learning_rate": 0.0021453619350789376, + "loss": 0.0808, + "num_input_tokens_seen": 49059968, + "step": 37845 + }, + { + "epoch": 1.8493880243324456, + "grad_norm": 0.0015081592136994004, + "learning_rate": 0.0021354464653955516, + "loss": 0.0853, + "num_input_tokens_seen": 49066208, + "step": 37850 + }, + { + "epoch": 1.84963232599614, + "grad_norm": 0.0014581253053620458, + "learning_rate": 0.002125553798284513, + "loss": 0.0982, + "num_input_tokens_seen": 49072512, + "step": 37855 + }, + { + "epoch": 1.8498766276598344, + "grad_norm": 0.00151222909335047, + "learning_rate": 0.002115683935271384, + "loss": 0.0748, + "num_input_tokens_seen": 49078848, + "step": 37860 + }, + { + "epoch": 1.8501209293235288, + "grad_norm": 0.001406384166330099, + "learning_rate": 0.0021058368778782144, + "loss": 0.0749, + "num_input_tokens_seen": 49085408, + "step": 37865 + }, + { + "epoch": 1.8503652309872232, + "grad_norm": 0.0013425133656710386, + "learning_rate": 0.002096012627623539, + "loss": 0.079, + "num_input_tokens_seen": 49091808, + "step": 37870 + }, + { + "epoch": 1.8506095326509173, + "grad_norm": 0.0023095495998859406, + "learning_rate": 0.00208621118602243, + "loss": 0.0937, + "num_input_tokens_seen": 49098176, + "step": 37875 + }, + { + "epoch": 1.8508538343146117, + "grad_norm": 0.0010345658520236611, + "learning_rate": 0.002076432554586327, + "loss": 0.0735, + "num_input_tokens_seen": 49104704, + "step": 37880 + }, + { + "epoch": 1.8510981359783059, + "grad_norm": 0.0015548382652923465, + "learning_rate": 0.002066676734823258, + "loss": 0.0978, + "num_input_tokens_seen": 49110752, + "step": 37885 + }, + { + "epoch": 1.8513424376420002, + "grad_norm": 0.0018998105078935623, + "learning_rate": 0.0020569437282376866, + "loss": 0.0834, + "num_input_tokens_seen": 49117792, + "step": 37890 + }, + { + "epoch": 1.8515867393056946, + "grad_norm": 0.0019924715161323547, + "learning_rate": 0.002047233536330545, + "loss": 0.0925, + "num_input_tokens_seen": 49124256, + "step": 37895 + }, + { + "epoch": 1.851831040969389, + "grad_norm": 0.002017814898863435, + "learning_rate": 0.0020375461605993015, + "loss": 0.0785, + "num_input_tokens_seen": 49130496, + "step": 37900 + }, + { + "epoch": 1.8520753426330834, + "grad_norm": 0.0010339542059227824, + "learning_rate": 0.002027881602537845, + "loss": 0.0615, + "num_input_tokens_seen": 49136768, + "step": 37905 + }, + { + "epoch": 1.8523196442967778, + "grad_norm": 0.0014499793760478497, + "learning_rate": 0.002018239863636567, + "loss": 0.1114, + "num_input_tokens_seen": 49143136, + "step": 37910 + }, + { + "epoch": 1.8525639459604721, + "grad_norm": 0.0011835724581032991, + "learning_rate": 0.002008620945382378, + "loss": 0.05, + "num_input_tokens_seen": 49149664, + "step": 37915 + }, + { + "epoch": 1.8528082476241663, + "grad_norm": 0.0012010436039417982, + "learning_rate": 0.001999024849258607, + "loss": 0.0747, + "num_input_tokens_seen": 49156128, + "step": 37920 + }, + { + "epoch": 1.8530525492878607, + "grad_norm": 0.000583147571887821, + "learning_rate": 0.001989451576745105, + "loss": 0.0611, + "num_input_tokens_seen": 49162784, + "step": 37925 + }, + { + "epoch": 1.8532968509515548, + "grad_norm": 0.0018235293682664633, + "learning_rate": 0.00197990112931819, + "loss": 0.0947, + "num_input_tokens_seen": 49169408, + "step": 37930 + }, + { + "epoch": 1.8535411526152492, + "grad_norm": 0.003022255841642618, + "learning_rate": 0.0019703735084506345, + "loss": 0.1057, + "num_input_tokens_seen": 49175872, + "step": 37935 + }, + { + "epoch": 1.8537854542789436, + "grad_norm": 0.0009797390084713697, + "learning_rate": 0.001960868715611763, + "loss": 0.0767, + "num_input_tokens_seen": 49182432, + "step": 37940 + }, + { + "epoch": 1.854029755942638, + "grad_norm": 0.0023832693696022034, + "learning_rate": 0.0019513867522673034, + "loss": 0.1103, + "num_input_tokens_seen": 49188448, + "step": 37945 + }, + { + "epoch": 1.8542740576063323, + "grad_norm": 0.0015063851606100798, + "learning_rate": 0.001941927619879502, + "loss": 0.0812, + "num_input_tokens_seen": 49194944, + "step": 37950 + }, + { + "epoch": 1.8545183592700267, + "grad_norm": 0.001905674347653985, + "learning_rate": 0.0019324913199070758, + "loss": 0.1032, + "num_input_tokens_seen": 49201280, + "step": 37955 + }, + { + "epoch": 1.854762660933721, + "grad_norm": 0.0016450562980026007, + "learning_rate": 0.0019230778538052106, + "loss": 0.065, + "num_input_tokens_seen": 49207712, + "step": 37960 + }, + { + "epoch": 1.8550069625974153, + "grad_norm": 0.000801712041720748, + "learning_rate": 0.0019136872230255952, + "loss": 0.0773, + "num_input_tokens_seen": 49214048, + "step": 37965 + }, + { + "epoch": 1.8552512642611096, + "grad_norm": 0.00141714361961931, + "learning_rate": 0.0019043194290164045, + "loss": 0.085, + "num_input_tokens_seen": 49220928, + "step": 37970 + }, + { + "epoch": 1.8554955659248038, + "grad_norm": 0.0014251713873818517, + "learning_rate": 0.0018949744732222162, + "loss": 0.0771, + "num_input_tokens_seen": 49227200, + "step": 37975 + }, + { + "epoch": 1.8557398675884982, + "grad_norm": 0.0020123699214309454, + "learning_rate": 0.0018856523570841776, + "loss": 0.0741, + "num_input_tokens_seen": 49233856, + "step": 37980 + }, + { + "epoch": 1.8559841692521926, + "grad_norm": 0.0012601427733898163, + "learning_rate": 0.0018763530820398555, + "loss": 0.1106, + "num_input_tokens_seen": 49240224, + "step": 37985 + }, + { + "epoch": 1.856228470915887, + "grad_norm": 0.0013504078378900886, + "learning_rate": 0.0018670766495233525, + "loss": 0.0548, + "num_input_tokens_seen": 49246592, + "step": 37990 + }, + { + "epoch": 1.8564727725795813, + "grad_norm": 0.001973886741325259, + "learning_rate": 0.001857823060965158, + "loss": 0.0849, + "num_input_tokens_seen": 49253088, + "step": 37995 + }, + { + "epoch": 1.8567170742432757, + "grad_norm": 0.0018966930219903588, + "learning_rate": 0.0018485923177923467, + "loss": 0.0914, + "num_input_tokens_seen": 49259584, + "step": 38000 + }, + { + "epoch": 1.8567170742432757, + "eval_loss": 0.08342055231332779, + "eval_runtime": 402.4657, + "eval_samples_per_second": 90.405, + "eval_steps_per_second": 22.603, + "num_input_tokens_seen": 49259584, + "step": 38000 + }, + { + "epoch": 1.85696137590697, + "grad_norm": 0.0009876586263999343, + "learning_rate": 0.001839384421428364, + "loss": 0.0824, + "num_input_tokens_seen": 49266720, + "step": 38005 + }, + { + "epoch": 1.8572056775706642, + "grad_norm": 0.0007899003685452044, + "learning_rate": 0.0018301993732932065, + "loss": 0.0957, + "num_input_tokens_seen": 49273248, + "step": 38010 + }, + { + "epoch": 1.8574499792343586, + "grad_norm": 0.0010044953087344766, + "learning_rate": 0.0018210371748033248, + "loss": 0.103, + "num_input_tokens_seen": 49279680, + "step": 38015 + }, + { + "epoch": 1.8576942808980528, + "grad_norm": 0.0015213723527267575, + "learning_rate": 0.0018118978273716556, + "loss": 0.0897, + "num_input_tokens_seen": 49286368, + "step": 38020 + }, + { + "epoch": 1.8579385825617472, + "grad_norm": 0.0022331736981868744, + "learning_rate": 0.001802781332407588, + "loss": 0.08, + "num_input_tokens_seen": 49292416, + "step": 38025 + }, + { + "epoch": 1.8581828842254415, + "grad_norm": 0.0009136634762398899, + "learning_rate": 0.0017936876913169806, + "loss": 0.1053, + "num_input_tokens_seen": 49298848, + "step": 38030 + }, + { + "epoch": 1.858427185889136, + "grad_norm": 0.0014903490664437413, + "learning_rate": 0.0017846169055022287, + "loss": 0.1035, + "num_input_tokens_seen": 49304864, + "step": 38035 + }, + { + "epoch": 1.8586714875528303, + "grad_norm": 0.0012546794023364782, + "learning_rate": 0.0017755689763621295, + "loss": 0.0851, + "num_input_tokens_seen": 49311296, + "step": 38040 + }, + { + "epoch": 1.8589157892165247, + "grad_norm": 0.002528173616155982, + "learning_rate": 0.0017665439052920173, + "loss": 0.0899, + "num_input_tokens_seen": 49317792, + "step": 38045 + }, + { + "epoch": 1.859160090880219, + "grad_norm": 0.0014378451742231846, + "learning_rate": 0.0017575416936836286, + "loss": 0.0678, + "num_input_tokens_seen": 49324320, + "step": 38050 + }, + { + "epoch": 1.8594043925439132, + "grad_norm": 0.002415244234725833, + "learning_rate": 0.0017485623429252528, + "loss": 0.0899, + "num_input_tokens_seen": 49331072, + "step": 38055 + }, + { + "epoch": 1.8596486942076076, + "grad_norm": 0.0021229602862149477, + "learning_rate": 0.0017396058544016156, + "loss": 0.0825, + "num_input_tokens_seen": 49337760, + "step": 38060 + }, + { + "epoch": 1.8598929958713017, + "grad_norm": 0.0012871926883235574, + "learning_rate": 0.0017306722294938958, + "loss": 0.0985, + "num_input_tokens_seen": 49343904, + "step": 38065 + }, + { + "epoch": 1.8601372975349961, + "grad_norm": 0.0014048685552552342, + "learning_rate": 0.0017217614695798078, + "loss": 0.0843, + "num_input_tokens_seen": 49350240, + "step": 38070 + }, + { + "epoch": 1.8603815991986905, + "grad_norm": 0.0008119090925902128, + "learning_rate": 0.001712873576033469, + "loss": 0.0792, + "num_input_tokens_seen": 49356480, + "step": 38075 + }, + { + "epoch": 1.8606259008623849, + "grad_norm": 0.0013576560886576772, + "learning_rate": 0.0017040085502255163, + "loss": 0.0645, + "num_input_tokens_seen": 49363104, + "step": 38080 + }, + { + "epoch": 1.8608702025260793, + "grad_norm": 0.0013443371281027794, + "learning_rate": 0.0016951663935230565, + "loss": 0.0588, + "num_input_tokens_seen": 49369504, + "step": 38085 + }, + { + "epoch": 1.8611145041897736, + "grad_norm": 0.0006518300506286323, + "learning_rate": 0.0016863471072896485, + "loss": 0.0869, + "num_input_tokens_seen": 49376288, + "step": 38090 + }, + { + "epoch": 1.861358805853468, + "grad_norm": 0.0018806994194164872, + "learning_rate": 0.0016775506928853377, + "loss": 0.0724, + "num_input_tokens_seen": 49382464, + "step": 38095 + }, + { + "epoch": 1.8616031075171622, + "grad_norm": 0.0015531014651060104, + "learning_rate": 0.001668777151666656, + "loss": 0.0661, + "num_input_tokens_seen": 49388864, + "step": 38100 + }, + { + "epoch": 1.8618474091808566, + "grad_norm": 0.000914559350349009, + "learning_rate": 0.0016600264849865709, + "loss": 0.088, + "num_input_tokens_seen": 49395168, + "step": 38105 + }, + { + "epoch": 1.8620917108445507, + "grad_norm": 0.0020967477466911077, + "learning_rate": 0.0016512986941945695, + "loss": 0.0793, + "num_input_tokens_seen": 49401536, + "step": 38110 + }, + { + "epoch": 1.862336012508245, + "grad_norm": 0.0018429995980113745, + "learning_rate": 0.0016425937806365753, + "loss": 0.0789, + "num_input_tokens_seen": 49407744, + "step": 38115 + }, + { + "epoch": 1.8625803141719395, + "grad_norm": 0.001189917791634798, + "learning_rate": 0.0016339117456549979, + "loss": 0.0613, + "num_input_tokens_seen": 49414912, + "step": 38120 + }, + { + "epoch": 1.8628246158356339, + "grad_norm": 0.0009840141283348203, + "learning_rate": 0.0016252525905886995, + "loss": 0.0859, + "num_input_tokens_seen": 49421280, + "step": 38125 + }, + { + "epoch": 1.8630689174993282, + "grad_norm": 0.001007694168947637, + "learning_rate": 0.0016166163167730617, + "loss": 0.076, + "num_input_tokens_seen": 49427232, + "step": 38130 + }, + { + "epoch": 1.8633132191630226, + "grad_norm": 0.0013443188508972526, + "learning_rate": 0.0016080029255398864, + "loss": 0.0807, + "num_input_tokens_seen": 49433792, + "step": 38135 + }, + { + "epoch": 1.863557520826717, + "grad_norm": 0.0022816972341388464, + "learning_rate": 0.0015994124182174606, + "loss": 0.0801, + "num_input_tokens_seen": 49440576, + "step": 38140 + }, + { + "epoch": 1.8638018224904112, + "grad_norm": 0.0010761291487142444, + "learning_rate": 0.001590844796130575, + "loss": 0.0814, + "num_input_tokens_seen": 49447232, + "step": 38145 + }, + { + "epoch": 1.8640461241541055, + "grad_norm": 0.0010329149663448334, + "learning_rate": 0.001582300060600439, + "loss": 0.0697, + "num_input_tokens_seen": 49453568, + "step": 38150 + }, + { + "epoch": 1.8642904258177997, + "grad_norm": 0.0017950218170881271, + "learning_rate": 0.0015737782129447652, + "loss": 0.0749, + "num_input_tokens_seen": 49460224, + "step": 38155 + }, + { + "epoch": 1.864534727481494, + "grad_norm": 0.0017641576705500484, + "learning_rate": 0.0015652792544777361, + "loss": 0.0841, + "num_input_tokens_seen": 49466688, + "step": 38160 + }, + { + "epoch": 1.8647790291451884, + "grad_norm": 0.001120561733841896, + "learning_rate": 0.0015568031865099863, + "loss": 0.0688, + "num_input_tokens_seen": 49472928, + "step": 38165 + }, + { + "epoch": 1.8650233308088828, + "grad_norm": 0.0021810720209032297, + "learning_rate": 0.0015483500103486369, + "loss": 0.0861, + "num_input_tokens_seen": 49479360, + "step": 38170 + }, + { + "epoch": 1.8652676324725772, + "grad_norm": 0.001365017262287438, + "learning_rate": 0.0015399197272972787, + "loss": 0.0696, + "num_input_tokens_seen": 49485856, + "step": 38175 + }, + { + "epoch": 1.8655119341362716, + "grad_norm": 0.0013992412714287639, + "learning_rate": 0.0015315123386559714, + "loss": 0.0839, + "num_input_tokens_seen": 49492064, + "step": 38180 + }, + { + "epoch": 1.8657562357999657, + "grad_norm": 0.0020472430624067783, + "learning_rate": 0.0015231278457212283, + "loss": 0.0981, + "num_input_tokens_seen": 49498848, + "step": 38185 + }, + { + "epoch": 1.8660005374636601, + "grad_norm": 0.0011813808232545853, + "learning_rate": 0.001514766249786048, + "loss": 0.0812, + "num_input_tokens_seen": 49505376, + "step": 38190 + }, + { + "epoch": 1.8662448391273545, + "grad_norm": 0.0011209185468032956, + "learning_rate": 0.0015064275521398994, + "loss": 0.0726, + "num_input_tokens_seen": 49511744, + "step": 38195 + }, + { + "epoch": 1.8664891407910487, + "grad_norm": 0.001150498166680336, + "learning_rate": 0.0014981117540686872, + "loss": 0.0747, + "num_input_tokens_seen": 49518144, + "step": 38200 + }, + { + "epoch": 1.8664891407910487, + "eval_loss": 0.08351525664329529, + "eval_runtime": 402.3344, + "eval_samples_per_second": 90.435, + "eval_steps_per_second": 22.611, + "num_input_tokens_seen": 49518144, + "step": 38200 + }, + { + "epoch": 1.866733442454743, + "grad_norm": 0.0022722689900547266, + "learning_rate": 0.0014898188568548687, + "loss": 0.0888, + "num_input_tokens_seen": 49524608, + "step": 38205 + }, + { + "epoch": 1.8669777441184374, + "grad_norm": 0.0027043477166444063, + "learning_rate": 0.0014815488617772542, + "loss": 0.1018, + "num_input_tokens_seen": 49531200, + "step": 38210 + }, + { + "epoch": 1.8672220457821318, + "grad_norm": 0.001018694369122386, + "learning_rate": 0.0014733017701112072, + "loss": 0.0972, + "num_input_tokens_seen": 49538144, + "step": 38215 + }, + { + "epoch": 1.8674663474458262, + "grad_norm": 0.0017282041953876615, + "learning_rate": 0.0014650775831285435, + "loss": 0.1165, + "num_input_tokens_seen": 49544544, + "step": 38220 + }, + { + "epoch": 1.8677106491095206, + "grad_norm": 0.0015815278748050332, + "learning_rate": 0.001456876302097515, + "loss": 0.0854, + "num_input_tokens_seen": 49550848, + "step": 38225 + }, + { + "epoch": 1.8679549507732147, + "grad_norm": 0.0009715078049339354, + "learning_rate": 0.0014486979282828604, + "loss": 0.0816, + "num_input_tokens_seen": 49557344, + "step": 38230 + }, + { + "epoch": 1.868199252436909, + "grad_norm": 0.0015724957920610905, + "learning_rate": 0.001440542462945804, + "loss": 0.0712, + "num_input_tokens_seen": 49563968, + "step": 38235 + }, + { + "epoch": 1.8684435541006035, + "grad_norm": 0.0016188949812203646, + "learning_rate": 0.0014324099073440232, + "loss": 0.088, + "num_input_tokens_seen": 49570336, + "step": 38240 + }, + { + "epoch": 1.8686878557642976, + "grad_norm": 0.0011347931576892734, + "learning_rate": 0.0014243002627316482, + "loss": 0.0884, + "num_input_tokens_seen": 49577024, + "step": 38245 + }, + { + "epoch": 1.868932157427992, + "grad_norm": 0.0017640285659581423, + "learning_rate": 0.0014162135303592781, + "loss": 0.116, + "num_input_tokens_seen": 49583360, + "step": 38250 + }, + { + "epoch": 1.8691764590916864, + "grad_norm": 0.0014283705968409777, + "learning_rate": 0.001408149711474016, + "loss": 0.0775, + "num_input_tokens_seen": 49590240, + "step": 38255 + }, + { + "epoch": 1.8694207607553808, + "grad_norm": 0.0008406831184402108, + "learning_rate": 0.0014001088073193834, + "loss": 0.0904, + "num_input_tokens_seen": 49596928, + "step": 38260 + }, + { + "epoch": 1.8696650624190752, + "grad_norm": 0.0010988347930833697, + "learning_rate": 0.0013920908191354052, + "loss": 0.0708, + "num_input_tokens_seen": 49603360, + "step": 38265 + }, + { + "epoch": 1.8699093640827695, + "grad_norm": 0.0024700183421373367, + "learning_rate": 0.001384095748158526, + "loss": 0.0838, + "num_input_tokens_seen": 49609536, + "step": 38270 + }, + { + "epoch": 1.8701536657464637, + "grad_norm": 0.0016307384939864278, + "learning_rate": 0.0013761235956217255, + "loss": 0.0827, + "num_input_tokens_seen": 49615744, + "step": 38275 + }, + { + "epoch": 1.870397967410158, + "grad_norm": 0.0016526717226952314, + "learning_rate": 0.0013681743627543873, + "loss": 0.0938, + "num_input_tokens_seen": 49621952, + "step": 38280 + }, + { + "epoch": 1.8706422690738524, + "grad_norm": 0.001021229662001133, + "learning_rate": 0.001360248050782381, + "loss": 0.0703, + "num_input_tokens_seen": 49628096, + "step": 38285 + }, + { + "epoch": 1.8708865707375466, + "grad_norm": 0.0013632795307785273, + "learning_rate": 0.001352344660928062, + "loss": 0.0893, + "num_input_tokens_seen": 49634784, + "step": 38290 + }, + { + "epoch": 1.871130872401241, + "grad_norm": 0.0012511530658230186, + "learning_rate": 0.0013444641944102052, + "loss": 0.0664, + "num_input_tokens_seen": 49640832, + "step": 38295 + }, + { + "epoch": 1.8713751740649354, + "grad_norm": 0.0007412682753056288, + "learning_rate": 0.0013366066524441056, + "loss": 0.0775, + "num_input_tokens_seen": 49647456, + "step": 38300 + }, + { + "epoch": 1.8716194757286297, + "grad_norm": 0.0007443453068844974, + "learning_rate": 0.0013287720362414768, + "loss": 0.0671, + "num_input_tokens_seen": 49654144, + "step": 38305 + }, + { + "epoch": 1.8718637773923241, + "grad_norm": 0.0010112863965332508, + "learning_rate": 0.0013209603470105025, + "loss": 0.0834, + "num_input_tokens_seen": 49660608, + "step": 38310 + }, + { + "epoch": 1.8721080790560185, + "grad_norm": 0.0014837377239018679, + "learning_rate": 0.0013131715859558857, + "loss": 0.084, + "num_input_tokens_seen": 49667264, + "step": 38315 + }, + { + "epoch": 1.8723523807197127, + "grad_norm": 0.0008410039590671659, + "learning_rate": 0.001305405754278699, + "loss": 0.0762, + "num_input_tokens_seen": 49673376, + "step": 38320 + }, + { + "epoch": 1.872596682383407, + "grad_norm": 0.0006797353853471577, + "learning_rate": 0.0012976628531765843, + "loss": 0.0738, + "num_input_tokens_seen": 49679488, + "step": 38325 + }, + { + "epoch": 1.8728409840471012, + "grad_norm": 0.0010886933887377381, + "learning_rate": 0.0012899428838435533, + "loss": 0.0954, + "num_input_tokens_seen": 49685632, + "step": 38330 + }, + { + "epoch": 1.8730852857107956, + "grad_norm": 0.0009483877802267671, + "learning_rate": 0.001282245847470137, + "loss": 0.0981, + "num_input_tokens_seen": 49691648, + "step": 38335 + }, + { + "epoch": 1.87332958737449, + "grad_norm": 0.0009772686753422022, + "learning_rate": 0.001274571745243319, + "loss": 0.0772, + "num_input_tokens_seen": 49698016, + "step": 38340 + }, + { + "epoch": 1.8735738890381843, + "grad_norm": 0.0010439205216243863, + "learning_rate": 0.0012669205783465364, + "loss": 0.0599, + "num_input_tokens_seen": 49704384, + "step": 38345 + }, + { + "epoch": 1.8738181907018787, + "grad_norm": 0.0017892169998958707, + "learning_rate": 0.001259292347959695, + "loss": 0.0812, + "num_input_tokens_seen": 49710880, + "step": 38350 + }, + { + "epoch": 1.874062492365573, + "grad_norm": 0.001252001035027206, + "learning_rate": 0.0012516870552591707, + "loss": 0.0787, + "num_input_tokens_seen": 49717216, + "step": 38355 + }, + { + "epoch": 1.8743067940292675, + "grad_norm": 0.0015551429241895676, + "learning_rate": 0.001244104701417792, + "loss": 0.077, + "num_input_tokens_seen": 49723712, + "step": 38360 + }, + { + "epoch": 1.8745510956929616, + "grad_norm": 0.001704273745417595, + "learning_rate": 0.0012365452876048565, + "loss": 0.0866, + "num_input_tokens_seen": 49730272, + "step": 38365 + }, + { + "epoch": 1.874795397356656, + "grad_norm": 0.0012875802349299192, + "learning_rate": 0.001229008814986099, + "loss": 0.0812, + "num_input_tokens_seen": 49736640, + "step": 38370 + }, + { + "epoch": 1.8750396990203502, + "grad_norm": 0.002072545699775219, + "learning_rate": 0.0012214952847237725, + "loss": 0.0787, + "num_input_tokens_seen": 49743008, + "step": 38375 + }, + { + "epoch": 1.8752840006840445, + "grad_norm": 0.002432707929983735, + "learning_rate": 0.0012140046979765339, + "loss": 0.1118, + "num_input_tokens_seen": 49749408, + "step": 38380 + }, + { + "epoch": 1.875528302347739, + "grad_norm": 0.001027593039907515, + "learning_rate": 0.0012065370558995258, + "loss": 0.0702, + "num_input_tokens_seen": 49756288, + "step": 38385 + }, + { + "epoch": 1.8757726040114333, + "grad_norm": 0.001235651783645153, + "learning_rate": 0.0011990923596443602, + "loss": 0.0602, + "num_input_tokens_seen": 49763104, + "step": 38390 + }, + { + "epoch": 1.8760169056751277, + "grad_norm": 0.0017129236366599798, + "learning_rate": 0.001191670610359119, + "loss": 0.0762, + "num_input_tokens_seen": 49769376, + "step": 38395 + }, + { + "epoch": 1.876261207338822, + "grad_norm": 0.001467698602937162, + "learning_rate": 0.0011842718091882865, + "loss": 0.0847, + "num_input_tokens_seen": 49775776, + "step": 38400 + }, + { + "epoch": 1.876261207338822, + "eval_loss": 0.08336079865694046, + "eval_runtime": 401.8571, + "eval_samples_per_second": 90.542, + "eval_steps_per_second": 22.637, + "num_input_tokens_seen": 49775776, + "step": 38400 + }, + { + "epoch": 1.8765055090025164, + "grad_norm": 0.0012793041532859206, + "learning_rate": 0.0011768959572729, + "loss": 0.0873, + "num_input_tokens_seen": 49782016, + "step": 38405 + }, + { + "epoch": 1.8767498106662106, + "grad_norm": 0.0009660903015173972, + "learning_rate": 0.001169543055750366, + "loss": 0.0757, + "num_input_tokens_seen": 49788448, + "step": 38410 + }, + { + "epoch": 1.876994112329905, + "grad_norm": 0.0022102538496255875, + "learning_rate": 0.0011622131057546115, + "loss": 0.0894, + "num_input_tokens_seen": 49795072, + "step": 38415 + }, + { + "epoch": 1.8772384139935991, + "grad_norm": 0.0014499177923426032, + "learning_rate": 0.0011549061084160316, + "loss": 0.108, + "num_input_tokens_seen": 49801376, + "step": 38420 + }, + { + "epoch": 1.8774827156572935, + "grad_norm": 0.0013569273287430406, + "learning_rate": 0.0011476220648614088, + "loss": 0.0965, + "num_input_tokens_seen": 49807648, + "step": 38425 + }, + { + "epoch": 1.877727017320988, + "grad_norm": 0.0017800208879634738, + "learning_rate": 0.0011403609762140777, + "loss": 0.075, + "num_input_tokens_seen": 49814528, + "step": 38430 + }, + { + "epoch": 1.8779713189846823, + "grad_norm": 0.0014522639103233814, + "learning_rate": 0.0011331228435937756, + "loss": 0.0753, + "num_input_tokens_seen": 49821248, + "step": 38435 + }, + { + "epoch": 1.8782156206483767, + "grad_norm": 0.0012691058218479156, + "learning_rate": 0.0011259076681166935, + "loss": 0.0597, + "num_input_tokens_seen": 49827456, + "step": 38440 + }, + { + "epoch": 1.878459922312071, + "grad_norm": 0.0016518669435754418, + "learning_rate": 0.0011187154508955244, + "loss": 0.0895, + "num_input_tokens_seen": 49833952, + "step": 38445 + }, + { + "epoch": 1.8787042239757654, + "grad_norm": 0.0017551897326484323, + "learning_rate": 0.001111546193039381, + "loss": 0.0745, + "num_input_tokens_seen": 49840640, + "step": 38450 + }, + { + "epoch": 1.8789485256394596, + "grad_norm": 0.0019732152577489614, + "learning_rate": 0.0011043998956538792, + "loss": 0.0734, + "num_input_tokens_seen": 49847008, + "step": 38455 + }, + { + "epoch": 1.879192827303154, + "grad_norm": 0.0014223293401300907, + "learning_rate": 0.0010972765598410538, + "loss": 0.0868, + "num_input_tokens_seen": 49853376, + "step": 38460 + }, + { + "epoch": 1.8794371289668481, + "grad_norm": 0.002059915103018284, + "learning_rate": 0.0010901761866993931, + "loss": 0.0754, + "num_input_tokens_seen": 49859936, + "step": 38465 + }, + { + "epoch": 1.8796814306305425, + "grad_norm": 0.0023333800490945578, + "learning_rate": 0.0010830987773238876, + "loss": 0.0686, + "num_input_tokens_seen": 49866656, + "step": 38470 + }, + { + "epoch": 1.8799257322942369, + "grad_norm": 0.0021015028469264507, + "learning_rate": 0.0010760443328059644, + "loss": 0.1116, + "num_input_tokens_seen": 49872928, + "step": 38475 + }, + { + "epoch": 1.8801700339579313, + "grad_norm": 0.0018188826506957412, + "learning_rate": 0.001069012854233503, + "loss": 0.0982, + "num_input_tokens_seen": 49879456, + "step": 38480 + }, + { + "epoch": 1.8804143356216256, + "grad_norm": 0.0016130603617057204, + "learning_rate": 0.0010620043426908365, + "loss": 0.0807, + "num_input_tokens_seen": 49886496, + "step": 38485 + }, + { + "epoch": 1.88065863728532, + "grad_norm": 0.001605330966413021, + "learning_rate": 0.0010550187992587833, + "loss": 0.0701, + "num_input_tokens_seen": 49893312, + "step": 38490 + }, + { + "epoch": 1.8809029389490144, + "grad_norm": 0.001470236573368311, + "learning_rate": 0.0010480562250145653, + "loss": 0.0673, + "num_input_tokens_seen": 49899520, + "step": 38495 + }, + { + "epoch": 1.8811472406127085, + "grad_norm": 0.0023112983908504248, + "learning_rate": 0.0010411166210319567, + "loss": 0.0936, + "num_input_tokens_seen": 49906048, + "step": 38500 + }, + { + "epoch": 1.881391542276403, + "grad_norm": 0.0020312070846557617, + "learning_rate": 0.0010341999883810848, + "loss": 0.0602, + "num_input_tokens_seen": 49912288, + "step": 38505 + }, + { + "epoch": 1.881635843940097, + "grad_norm": 0.00169869523961097, + "learning_rate": 0.0010273063281285965, + "loss": 0.099, + "num_input_tokens_seen": 49919008, + "step": 38510 + }, + { + "epoch": 1.8818801456037915, + "grad_norm": 0.0011436032364144921, + "learning_rate": 0.0010204356413375747, + "loss": 0.1079, + "num_input_tokens_seen": 49925248, + "step": 38515 + }, + { + "epoch": 1.8821244472674858, + "grad_norm": 0.0018355746287852526, + "learning_rate": 0.001013587929067572, + "loss": 0.0658, + "num_input_tokens_seen": 49931296, + "step": 38520 + }, + { + "epoch": 1.8823687489311802, + "grad_norm": 0.0009352895431220531, + "learning_rate": 0.00100676319237461, + "loss": 0.0647, + "num_input_tokens_seen": 49937568, + "step": 38525 + }, + { + "epoch": 1.8826130505948746, + "grad_norm": 0.0014141614083200693, + "learning_rate": 0.0009999614323110972, + "loss": 0.083, + "num_input_tokens_seen": 49944544, + "step": 38530 + }, + { + "epoch": 1.882857352258569, + "grad_norm": 0.0008962914580479264, + "learning_rate": 0.000993182649926011, + "loss": 0.0705, + "num_input_tokens_seen": 49951392, + "step": 38535 + }, + { + "epoch": 1.8831016539222634, + "grad_norm": 0.0013049569679424167, + "learning_rate": 0.000986426846264682, + "loss": 0.0867, + "num_input_tokens_seen": 49957824, + "step": 38540 + }, + { + "epoch": 1.8833459555859575, + "grad_norm": 0.0016247910680249333, + "learning_rate": 0.00097969402236896, + "loss": 0.0723, + "num_input_tokens_seen": 49964640, + "step": 38545 + }, + { + "epoch": 1.883590257249652, + "grad_norm": 0.0018403180874884129, + "learning_rate": 0.0009729841792771143, + "loss": 0.0905, + "num_input_tokens_seen": 49971136, + "step": 38550 + }, + { + "epoch": 1.883834558913346, + "grad_norm": 0.0020664501935243607, + "learning_rate": 0.0009662973180239176, + "loss": 0.112, + "num_input_tokens_seen": 49977408, + "step": 38555 + }, + { + "epoch": 1.8840788605770404, + "grad_norm": 0.0019778453279286623, + "learning_rate": 0.0009596334396405448, + "loss": 0.0887, + "num_input_tokens_seen": 49983680, + "step": 38560 + }, + { + "epoch": 1.8843231622407348, + "grad_norm": 0.002230504062026739, + "learning_rate": 0.0009529925451546406, + "loss": 0.0691, + "num_input_tokens_seen": 49990144, + "step": 38565 + }, + { + "epoch": 1.8845674639044292, + "grad_norm": 0.0008984023588709533, + "learning_rate": 0.0009463746355903357, + "loss": 0.1012, + "num_input_tokens_seen": 49996640, + "step": 38570 + }, + { + "epoch": 1.8848117655681236, + "grad_norm": 0.0026623932644724846, + "learning_rate": 0.0009397797119681971, + "loss": 0.0916, + "num_input_tokens_seen": 50003296, + "step": 38575 + }, + { + "epoch": 1.885056067231818, + "grad_norm": 0.001529203960672021, + "learning_rate": 0.0009332077753052281, + "loss": 0.0747, + "num_input_tokens_seen": 50009920, + "step": 38580 + }, + { + "epoch": 1.8853003688955123, + "grad_norm": 0.0011919738026335835, + "learning_rate": 0.0009266588266149011, + "loss": 0.1066, + "num_input_tokens_seen": 50016672, + "step": 38585 + }, + { + "epoch": 1.8855446705592065, + "grad_norm": 0.0012661145301535726, + "learning_rate": 0.0009201328669071584, + "loss": 0.0718, + "num_input_tokens_seen": 50023072, + "step": 38590 + }, + { + "epoch": 1.8857889722229009, + "grad_norm": 0.0023974482901394367, + "learning_rate": 0.0009136298971883949, + "loss": 0.0648, + "num_input_tokens_seen": 50029600, + "step": 38595 + }, + { + "epoch": 1.886033273886595, + "grad_norm": 0.0012652986915782094, + "learning_rate": 0.0009071499184614251, + "loss": 0.0699, + "num_input_tokens_seen": 50036384, + "step": 38600 + }, + { + "epoch": 1.886033273886595, + "eval_loss": 0.08342919498682022, + "eval_runtime": 402.2484, + "eval_samples_per_second": 90.454, + "eval_steps_per_second": 22.615, + "num_input_tokens_seen": 50036384, + "step": 38600 + }, + { + "epoch": 1.8862775755502894, + "grad_norm": 0.001504500862210989, + "learning_rate": 0.0009006929317255663, + "loss": 0.0604, + "num_input_tokens_seen": 50042752, + "step": 38605 + }, + { + "epoch": 1.8865218772139838, + "grad_norm": 0.0008777306065894663, + "learning_rate": 0.0008942589379765387, + "loss": 0.0721, + "num_input_tokens_seen": 50048928, + "step": 38610 + }, + { + "epoch": 1.8867661788776782, + "grad_norm": 0.0008751774439588189, + "learning_rate": 0.0008878479382065817, + "loss": 0.0533, + "num_input_tokens_seen": 50055424, + "step": 38615 + }, + { + "epoch": 1.8870104805413725, + "grad_norm": 0.0016626009019091725, + "learning_rate": 0.0008814599334043215, + "loss": 0.0663, + "num_input_tokens_seen": 50061568, + "step": 38620 + }, + { + "epoch": 1.887254782205067, + "grad_norm": 0.001842924510128796, + "learning_rate": 0.0008750949245548866, + "loss": 0.0941, + "num_input_tokens_seen": 50067712, + "step": 38625 + }, + { + "epoch": 1.8874990838687613, + "grad_norm": 0.0013542131055146456, + "learning_rate": 0.0008687529126398252, + "loss": 0.0895, + "num_input_tokens_seen": 50073888, + "step": 38630 + }, + { + "epoch": 1.8877433855324555, + "grad_norm": 0.0014537409879267216, + "learning_rate": 0.0008624338986371715, + "loss": 0.0674, + "num_input_tokens_seen": 50080256, + "step": 38635 + }, + { + "epoch": 1.8879876871961498, + "grad_norm": 0.001206149929203093, + "learning_rate": 0.0008561378835213962, + "loss": 0.0796, + "num_input_tokens_seen": 50086752, + "step": 38640 + }, + { + "epoch": 1.888231988859844, + "grad_norm": 0.001394058228470385, + "learning_rate": 0.0008498648682634058, + "loss": 0.0816, + "num_input_tokens_seen": 50093632, + "step": 38645 + }, + { + "epoch": 1.8884762905235384, + "grad_norm": 0.0010956706246361136, + "learning_rate": 0.0008436148538306099, + "loss": 0.0595, + "num_input_tokens_seen": 50100736, + "step": 38650 + }, + { + "epoch": 1.8887205921872328, + "grad_norm": 0.0019996289629489183, + "learning_rate": 0.0008373878411868041, + "loss": 0.0905, + "num_input_tokens_seen": 50107232, + "step": 38655 + }, + { + "epoch": 1.8889648938509271, + "grad_norm": 0.0009042808087542653, + "learning_rate": 0.000831183831292287, + "loss": 0.072, + "num_input_tokens_seen": 50113728, + "step": 38660 + }, + { + "epoch": 1.8892091955146215, + "grad_norm": 0.0008238825830630958, + "learning_rate": 0.0008250028251037933, + "loss": 0.0864, + "num_input_tokens_seen": 50120192, + "step": 38665 + }, + { + "epoch": 1.889453497178316, + "grad_norm": 0.0008596939733251929, + "learning_rate": 0.0008188448235745271, + "loss": 0.0943, + "num_input_tokens_seen": 50126208, + "step": 38670 + }, + { + "epoch": 1.8896977988420103, + "grad_norm": 0.0012569722020998597, + "learning_rate": 0.0008127098276541122, + "loss": 0.0897, + "num_input_tokens_seen": 50132896, + "step": 38675 + }, + { + "epoch": 1.8899421005057044, + "grad_norm": 0.0012977187288925052, + "learning_rate": 0.0008065978382886418, + "loss": 0.09, + "num_input_tokens_seen": 50139328, + "step": 38680 + }, + { + "epoch": 1.8901864021693988, + "grad_norm": 0.0012436701217666268, + "learning_rate": 0.0008005088564206785, + "loss": 0.0924, + "num_input_tokens_seen": 50145216, + "step": 38685 + }, + { + "epoch": 1.890430703833093, + "grad_norm": 0.0008105356246232986, + "learning_rate": 0.0007944428829891881, + "loss": 0.0625, + "num_input_tokens_seen": 50152992, + "step": 38690 + }, + { + "epoch": 1.8906750054967874, + "grad_norm": 0.0012613266007974744, + "learning_rate": 0.0007883999189296386, + "loss": 0.0817, + "num_input_tokens_seen": 50159680, + "step": 38695 + }, + { + "epoch": 1.8909193071604817, + "grad_norm": 0.001651573576964438, + "learning_rate": 0.0007823799651739515, + "loss": 0.0987, + "num_input_tokens_seen": 50166368, + "step": 38700 + }, + { + "epoch": 1.8911636088241761, + "grad_norm": 0.0010180219542235136, + "learning_rate": 0.0007763830226504509, + "loss": 0.0741, + "num_input_tokens_seen": 50173024, + "step": 38705 + }, + { + "epoch": 1.8914079104878705, + "grad_norm": 0.0013151437742635608, + "learning_rate": 0.0007704090922839468, + "loss": 0.1156, + "num_input_tokens_seen": 50179456, + "step": 38710 + }, + { + "epoch": 1.8916522121515649, + "grad_norm": 0.0017500873655080795, + "learning_rate": 0.0007644581749957025, + "loss": 0.0881, + "num_input_tokens_seen": 50185920, + "step": 38715 + }, + { + "epoch": 1.891896513815259, + "grad_norm": 0.002512756735086441, + "learning_rate": 0.000758530271703417, + "loss": 0.0834, + "num_input_tokens_seen": 50192448, + "step": 38720 + }, + { + "epoch": 1.8921408154789534, + "grad_norm": 0.0018648153636604548, + "learning_rate": 0.0007526253833212426, + "loss": 0.0846, + "num_input_tokens_seen": 50199200, + "step": 38725 + }, + { + "epoch": 1.8923851171426478, + "grad_norm": 0.0011783173540607095, + "learning_rate": 0.0007467435107598008, + "loss": 0.0616, + "num_input_tokens_seen": 50205760, + "step": 38730 + }, + { + "epoch": 1.892629418806342, + "grad_norm": 0.0011781023349612951, + "learning_rate": 0.0007408846549261328, + "loss": 0.0768, + "num_input_tokens_seen": 50212352, + "step": 38735 + }, + { + "epoch": 1.8928737204700363, + "grad_norm": 0.0009322012774646282, + "learning_rate": 0.0007350488167237656, + "loss": 0.0759, + "num_input_tokens_seen": 50219232, + "step": 38740 + }, + { + "epoch": 1.8931180221337307, + "grad_norm": 0.0014702717307955027, + "learning_rate": 0.0007292359970526629, + "loss": 0.0843, + "num_input_tokens_seen": 50225440, + "step": 38745 + }, + { + "epoch": 1.893362323797425, + "grad_norm": 0.0016196670476347208, + "learning_rate": 0.0007234461968092076, + "loss": 0.0869, + "num_input_tokens_seen": 50232064, + "step": 38750 + }, + { + "epoch": 1.8936066254611195, + "grad_norm": 0.0017132462235167623, + "learning_rate": 0.0007176794168862854, + "loss": 0.1011, + "num_input_tokens_seen": 50238720, + "step": 38755 + }, + { + "epoch": 1.8938509271248138, + "grad_norm": 0.0011550731724128127, + "learning_rate": 0.000711935658173185, + "loss": 0.0746, + "num_input_tokens_seen": 50245440, + "step": 38760 + }, + { + "epoch": 1.894095228788508, + "grad_norm": 0.0011053637135773897, + "learning_rate": 0.0007062149215556812, + "loss": 0.0903, + "num_input_tokens_seen": 50251936, + "step": 38765 + }, + { + "epoch": 1.8943395304522024, + "grad_norm": 0.0009143630159087479, + "learning_rate": 0.0007005172079159849, + "loss": 0.0779, + "num_input_tokens_seen": 50258912, + "step": 38770 + }, + { + "epoch": 1.8945838321158968, + "grad_norm": 0.002229492412880063, + "learning_rate": 0.0006948425181327267, + "loss": 0.0839, + "num_input_tokens_seen": 50265600, + "step": 38775 + }, + { + "epoch": 1.894828133779591, + "grad_norm": 0.001350597944110632, + "learning_rate": 0.000689190853081073, + "loss": 0.0815, + "num_input_tokens_seen": 50272128, + "step": 38780 + }, + { + "epoch": 1.8950724354432853, + "grad_norm": 0.0009147960226982832, + "learning_rate": 0.000683562213632527, + "loss": 0.0787, + "num_input_tokens_seen": 50278720, + "step": 38785 + }, + { + "epoch": 1.8953167371069797, + "grad_norm": 0.0009264491382054985, + "learning_rate": 0.0006779566006551108, + "loss": 0.0595, + "num_input_tokens_seen": 50285504, + "step": 38790 + }, + { + "epoch": 1.895561038770674, + "grad_norm": 0.002339794300496578, + "learning_rate": 0.0006723740150132995, + "loss": 0.081, + "num_input_tokens_seen": 50292128, + "step": 38795 + }, + { + "epoch": 1.8958053404343684, + "grad_norm": 0.0012016583932563663, + "learning_rate": 0.0006668144575679713, + "loss": 0.0633, + "num_input_tokens_seen": 50298720, + "step": 38800 + }, + { + "epoch": 1.8958053404343684, + "eval_loss": 0.08346465975046158, + "eval_runtime": 401.7742, + "eval_samples_per_second": 90.561, + "eval_steps_per_second": 22.642, + "num_input_tokens_seen": 50298720, + "step": 38800 + }, + { + "epoch": 1.8960496420980628, + "grad_norm": 0.0015116985887289047, + "learning_rate": 0.0006612779291765069, + "loss": 0.0725, + "num_input_tokens_seen": 50305440, + "step": 38805 + }, + { + "epoch": 1.896293943761757, + "grad_norm": 0.0018948332872241735, + "learning_rate": 0.0006557644306926736, + "loss": 0.0727, + "num_input_tokens_seen": 50311936, + "step": 38810 + }, + { + "epoch": 1.8965382454254514, + "grad_norm": 0.0007842591730877757, + "learning_rate": 0.0006502739629667575, + "loss": 0.0788, + "num_input_tokens_seen": 50318432, + "step": 38815 + }, + { + "epoch": 1.8967825470891457, + "grad_norm": 0.0010888047982007265, + "learning_rate": 0.0006448065268454317, + "loss": 0.0826, + "num_input_tokens_seen": 50324928, + "step": 38820 + }, + { + "epoch": 1.89702684875284, + "grad_norm": 0.002588969422504306, + "learning_rate": 0.0006393621231718549, + "loss": 0.117, + "num_input_tokens_seen": 50331424, + "step": 38825 + }, + { + "epoch": 1.8972711504165343, + "grad_norm": 0.0019058637553825974, + "learning_rate": 0.0006339407527856389, + "loss": 0.0965, + "num_input_tokens_seen": 50337696, + "step": 38830 + }, + { + "epoch": 1.8975154520802286, + "grad_norm": 0.0013886464294046164, + "learning_rate": 0.0006285424165227982, + "loss": 0.0661, + "num_input_tokens_seen": 50344608, + "step": 38835 + }, + { + "epoch": 1.897759753743923, + "grad_norm": 0.0014158434933051467, + "learning_rate": 0.0006231671152158169, + "loss": 0.0677, + "num_input_tokens_seen": 50351904, + "step": 38840 + }, + { + "epoch": 1.8980040554076174, + "grad_norm": 0.001239896984770894, + "learning_rate": 0.0006178148496936819, + "loss": 0.0771, + "num_input_tokens_seen": 50357920, + "step": 38845 + }, + { + "epoch": 1.8982483570713118, + "grad_norm": 0.0013788114301860332, + "learning_rate": 0.000612485620781733, + "loss": 0.0752, + "num_input_tokens_seen": 50363936, + "step": 38850 + }, + { + "epoch": 1.898492658735006, + "grad_norm": 0.0012497827410697937, + "learning_rate": 0.0006071794293018296, + "loss": 0.0763, + "num_input_tokens_seen": 50370784, + "step": 38855 + }, + { + "epoch": 1.8987369603987003, + "grad_norm": 0.0016660415567457676, + "learning_rate": 0.0006018962760722501, + "loss": 0.1106, + "num_input_tokens_seen": 50376960, + "step": 38860 + }, + { + "epoch": 1.8989812620623945, + "grad_norm": 0.0017763037467375398, + "learning_rate": 0.0005966361619077098, + "loss": 0.0791, + "num_input_tokens_seen": 50383584, + "step": 38865 + }, + { + "epoch": 1.8992255637260889, + "grad_norm": 0.0013449377147480845, + "learning_rate": 0.000591399087619393, + "loss": 0.0903, + "num_input_tokens_seen": 50390496, + "step": 38870 + }, + { + "epoch": 1.8994698653897832, + "grad_norm": 0.002246580319479108, + "learning_rate": 0.0005861850540149371, + "loss": 0.1019, + "num_input_tokens_seen": 50396704, + "step": 38875 + }, + { + "epoch": 1.8997141670534776, + "grad_norm": 0.001256345771253109, + "learning_rate": 0.0005809940618983822, + "loss": 0.0869, + "num_input_tokens_seen": 50403264, + "step": 38880 + }, + { + "epoch": 1.899958468717172, + "grad_norm": 0.0015022734878584743, + "learning_rate": 0.0005758261120702712, + "loss": 0.0757, + "num_input_tokens_seen": 50410208, + "step": 38885 + }, + { + "epoch": 1.9002027703808664, + "grad_norm": 0.0006697920034639537, + "learning_rate": 0.0005706812053275501, + "loss": 0.0894, + "num_input_tokens_seen": 50416736, + "step": 38890 + }, + { + "epoch": 1.9004470720445608, + "grad_norm": 0.002598490798845887, + "learning_rate": 0.0005655593424636173, + "loss": 0.0893, + "num_input_tokens_seen": 50423360, + "step": 38895 + }, + { + "epoch": 1.900691373708255, + "grad_norm": 0.0014705253997817636, + "learning_rate": 0.0005604605242683746, + "loss": 0.1134, + "num_input_tokens_seen": 50430208, + "step": 38900 + }, + { + "epoch": 1.9009356753719493, + "grad_norm": 0.0010710915084928274, + "learning_rate": 0.0005553847515280596, + "loss": 0.0771, + "num_input_tokens_seen": 50436928, + "step": 38905 + }, + { + "epoch": 1.9011799770356435, + "grad_norm": 0.00158511265181005, + "learning_rate": 0.0005503320250254795, + "loss": 0.0645, + "num_input_tokens_seen": 50443104, + "step": 38910 + }, + { + "epoch": 1.9014242786993378, + "grad_norm": 0.0010551685700193048, + "learning_rate": 0.0005453023455397943, + "loss": 0.0653, + "num_input_tokens_seen": 50449824, + "step": 38915 + }, + { + "epoch": 1.9016685803630322, + "grad_norm": 0.0011935323709622025, + "learning_rate": 0.0005402957138466502, + "loss": 0.0567, + "num_input_tokens_seen": 50456640, + "step": 38920 + }, + { + "epoch": 1.9019128820267266, + "grad_norm": 0.0024107072968035936, + "learning_rate": 0.0005353121307181463, + "loss": 0.087, + "num_input_tokens_seen": 50463488, + "step": 38925 + }, + { + "epoch": 1.902157183690421, + "grad_norm": 0.0014778648037463427, + "learning_rate": 0.0005303515969227845, + "loss": 0.0926, + "num_input_tokens_seen": 50469664, + "step": 38930 + }, + { + "epoch": 1.9024014853541154, + "grad_norm": 0.0011161259608343244, + "learning_rate": 0.0005254141132255862, + "loss": 0.0862, + "num_input_tokens_seen": 50476288, + "step": 38935 + }, + { + "epoch": 1.9026457870178097, + "grad_norm": 0.0021839241962879896, + "learning_rate": 0.0005204996803879258, + "loss": 0.0888, + "num_input_tokens_seen": 50482688, + "step": 38940 + }, + { + "epoch": 1.9028900886815039, + "grad_norm": 0.0012512897374108434, + "learning_rate": 0.0005156082991676969, + "loss": 0.0877, + "num_input_tokens_seen": 50488960, + "step": 38945 + }, + { + "epoch": 1.9031343903451983, + "grad_norm": 0.000992674264125526, + "learning_rate": 0.0005107399703192127, + "loss": 0.064, + "num_input_tokens_seen": 50495488, + "step": 38950 + }, + { + "epoch": 1.9033786920088924, + "grad_norm": 0.0010819289600476623, + "learning_rate": 0.0005058946945932063, + "loss": 0.0925, + "num_input_tokens_seen": 50502336, + "step": 38955 + }, + { + "epoch": 1.9036229936725868, + "grad_norm": 0.0009653886663727462, + "learning_rate": 0.0005010724727369131, + "loss": 0.0671, + "num_input_tokens_seen": 50508576, + "step": 38960 + }, + { + "epoch": 1.9038672953362812, + "grad_norm": 0.0010961830848827958, + "learning_rate": 0.000496273305493955, + "loss": 0.0769, + "num_input_tokens_seen": 50515232, + "step": 38965 + }, + { + "epoch": 1.9041115969999756, + "grad_norm": 0.0013858421007171273, + "learning_rate": 0.0004914971936044399, + "loss": 0.0799, + "num_input_tokens_seen": 50521376, + "step": 38970 + }, + { + "epoch": 1.90435589866367, + "grad_norm": 0.0013842876069247723, + "learning_rate": 0.00048674413780491196, + "loss": 0.0752, + "num_input_tokens_seen": 50527904, + "step": 38975 + }, + { + "epoch": 1.9046002003273643, + "grad_norm": 0.0015727211721241474, + "learning_rate": 0.0004820141388283183, + "loss": 0.0809, + "num_input_tokens_seen": 50534592, + "step": 38980 + }, + { + "epoch": 1.9048445019910587, + "grad_norm": 0.0010753544047474861, + "learning_rate": 0.00047730719740410874, + "loss": 0.0808, + "num_input_tokens_seen": 50541088, + "step": 38985 + }, + { + "epoch": 1.9050888036547529, + "grad_norm": 0.0012616509338840842, + "learning_rate": 0.00047262331425816927, + "loss": 0.0623, + "num_input_tokens_seen": 50547488, + "step": 38990 + }, + { + "epoch": 1.9053331053184472, + "grad_norm": 0.0011702267220243812, + "learning_rate": 0.00046796249011277213, + "loss": 0.0906, + "num_input_tokens_seen": 50554112, + "step": 38995 + }, + { + "epoch": 1.9055774069821414, + "grad_norm": 0.0012160316109657288, + "learning_rate": 0.00046332472568669236, + "loss": 0.082, + "num_input_tokens_seen": 50560704, + "step": 39000 + }, + { + "epoch": 1.9055774069821414, + "eval_loss": 0.08352769911289215, + "eval_runtime": 402.2526, + "eval_samples_per_second": 90.453, + "eval_steps_per_second": 22.615, + "num_input_tokens_seen": 50560704, + "step": 39000 + }, + { + "epoch": 1.9058217086458358, + "grad_norm": 0.0016057598404586315, + "learning_rate": 0.0004587100216951578, + "loss": 0.0907, + "num_input_tokens_seen": 50567296, + "step": 39005 + }, + { + "epoch": 1.9060660103095302, + "grad_norm": 0.001419192529283464, + "learning_rate": 0.00045411837884978265, + "loss": 0.1129, + "num_input_tokens_seen": 50573888, + "step": 39010 + }, + { + "epoch": 1.9063103119732245, + "grad_norm": 0.001009492320008576, + "learning_rate": 0.00044954979785865045, + "loss": 0.0677, + "num_input_tokens_seen": 50580192, + "step": 39015 + }, + { + "epoch": 1.906554613636919, + "grad_norm": 0.0021873617079108953, + "learning_rate": 0.00044500427942631426, + "loss": 0.0964, + "num_input_tokens_seen": 50586432, + "step": 39020 + }, + { + "epoch": 1.9067989153006133, + "grad_norm": 0.0011966880410909653, + "learning_rate": 0.0004404818242537467, + "loss": 0.0481, + "num_input_tokens_seen": 50593120, + "step": 39025 + }, + { + "epoch": 1.9070432169643077, + "grad_norm": 0.00128910131752491, + "learning_rate": 0.00043598243303837324, + "loss": 0.0936, + "num_input_tokens_seen": 50599264, + "step": 39030 + }, + { + "epoch": 1.9072875186280018, + "grad_norm": 0.0017264753114432096, + "learning_rate": 0.00043150610647403885, + "loss": 0.0635, + "num_input_tokens_seen": 50605952, + "step": 39035 + }, + { + "epoch": 1.9075318202916962, + "grad_norm": 0.0011551899369806051, + "learning_rate": 0.00042705284525104134, + "loss": 0.0598, + "num_input_tokens_seen": 50612576, + "step": 39040 + }, + { + "epoch": 1.9077761219553904, + "grad_norm": 0.0010033752769231796, + "learning_rate": 0.0004226226500561647, + "loss": 0.0792, + "num_input_tokens_seen": 50618880, + "step": 39045 + }, + { + "epoch": 1.9080204236190847, + "grad_norm": 0.001966457348316908, + "learning_rate": 0.0004182155215725791, + "loss": 0.0756, + "num_input_tokens_seen": 50625280, + "step": 39050 + }, + { + "epoch": 1.9082647252827791, + "grad_norm": 0.001771910465322435, + "learning_rate": 0.00041383146047992424, + "loss": 0.1037, + "num_input_tokens_seen": 50631456, + "step": 39055 + }, + { + "epoch": 1.9085090269464735, + "grad_norm": 0.0018967220094054937, + "learning_rate": 0.00040947046745427597, + "loss": 0.0649, + "num_input_tokens_seen": 50638048, + "step": 39060 + }, + { + "epoch": 1.9087533286101679, + "grad_norm": 0.0016959296772256494, + "learning_rate": 0.00040513254316814625, + "loss": 0.1166, + "num_input_tokens_seen": 50644800, + "step": 39065 + }, + { + "epoch": 1.9089976302738623, + "grad_norm": 0.0015253580641001463, + "learning_rate": 0.0004008176882905168, + "loss": 0.0851, + "num_input_tokens_seen": 50651008, + "step": 39070 + }, + { + "epoch": 1.9092419319375566, + "grad_norm": 0.0013509616255760193, + "learning_rate": 0.00039652590348677184, + "loss": 0.0784, + "num_input_tokens_seen": 50657312, + "step": 39075 + }, + { + "epoch": 1.9094862336012508, + "grad_norm": 0.0008953963406383991, + "learning_rate": 0.00039225718941878206, + "loss": 0.0623, + "num_input_tokens_seen": 50663872, + "step": 39080 + }, + { + "epoch": 1.9097305352649452, + "grad_norm": 0.0010023227659985423, + "learning_rate": 0.00038801154674480417, + "loss": 0.0962, + "num_input_tokens_seen": 50670752, + "step": 39085 + }, + { + "epoch": 1.9099748369286393, + "grad_norm": 0.0015580648323521018, + "learning_rate": 0.00038378897611959784, + "loss": 0.0977, + "num_input_tokens_seen": 50677184, + "step": 39090 + }, + { + "epoch": 1.9102191385923337, + "grad_norm": 0.002162869554013014, + "learning_rate": 0.00037958947819430875, + "loss": 0.1204, + "num_input_tokens_seen": 50684128, + "step": 39095 + }, + { + "epoch": 1.910463440256028, + "grad_norm": 0.0009535101125948131, + "learning_rate": 0.0003754130536165856, + "loss": 0.0959, + "num_input_tokens_seen": 50690752, + "step": 39100 + }, + { + "epoch": 1.9107077419197225, + "grad_norm": 0.0016965123359113932, + "learning_rate": 0.0003712597030304632, + "loss": 0.0965, + "num_input_tokens_seen": 50697504, + "step": 39105 + }, + { + "epoch": 1.9109520435834169, + "grad_norm": 0.0008283084607683122, + "learning_rate": 0.00036712942707646247, + "loss": 0.0457, + "num_input_tokens_seen": 50704480, + "step": 39110 + }, + { + "epoch": 1.9111963452471112, + "grad_norm": 0.0013071782886981964, + "learning_rate": 0.00036302222639149063, + "loss": 0.0726, + "num_input_tokens_seen": 50710944, + "step": 39115 + }, + { + "epoch": 1.9114406469108056, + "grad_norm": 0.0010700508719310164, + "learning_rate": 0.000358938101608941, + "loss": 0.0536, + "num_input_tokens_seen": 50716896, + "step": 39120 + }, + { + "epoch": 1.9116849485744998, + "grad_norm": 0.001290413667447865, + "learning_rate": 0.0003548770533586598, + "loss": 0.094, + "num_input_tokens_seen": 50723424, + "step": 39125 + }, + { + "epoch": 1.9119292502381942, + "grad_norm": 0.002023292239755392, + "learning_rate": 0.0003508390822668961, + "loss": 0.0859, + "num_input_tokens_seen": 50729760, + "step": 39130 + }, + { + "epoch": 1.9121735519018883, + "grad_norm": 0.0006425651954486966, + "learning_rate": 0.00034682418895633503, + "loss": 0.0729, + "num_input_tokens_seen": 50736416, + "step": 39135 + }, + { + "epoch": 1.9124178535655827, + "grad_norm": 0.0007187841110862792, + "learning_rate": 0.0003428323740461647, + "loss": 0.0785, + "num_input_tokens_seen": 50743168, + "step": 39140 + }, + { + "epoch": 1.912662155229277, + "grad_norm": 0.0013909941771999002, + "learning_rate": 0.00033886363815194276, + "loss": 0.1134, + "num_input_tokens_seen": 50749344, + "step": 39145 + }, + { + "epoch": 1.9129064568929715, + "grad_norm": 0.001394927967339754, + "learning_rate": 0.0003349179818857129, + "loss": 0.095, + "num_input_tokens_seen": 50755744, + "step": 39150 + }, + { + "epoch": 1.9131507585566658, + "grad_norm": 0.0012980102328583598, + "learning_rate": 0.0003309954058559383, + "loss": 0.0742, + "num_input_tokens_seen": 50761792, + "step": 39155 + }, + { + "epoch": 1.9133950602203602, + "grad_norm": 0.0016450684051960707, + "learning_rate": 0.0003270959106675186, + "loss": 0.076, + "num_input_tokens_seen": 50768192, + "step": 39160 + }, + { + "epoch": 1.9136393618840546, + "grad_norm": 0.001098791602998972, + "learning_rate": 0.0003232194969218227, + "loss": 0.092, + "num_input_tokens_seen": 50774496, + "step": 39165 + }, + { + "epoch": 1.9138836635477487, + "grad_norm": 0.0011655592825263739, + "learning_rate": 0.00031936616521663905, + "loss": 0.0756, + "num_input_tokens_seen": 50781120, + "step": 39170 + }, + { + "epoch": 1.9141279652114431, + "grad_norm": 0.001157216727733612, + "learning_rate": 0.00031553591614619236, + "loss": 0.0938, + "num_input_tokens_seen": 50787872, + "step": 39175 + }, + { + "epoch": 1.9143722668751373, + "grad_norm": 0.0017004822148010135, + "learning_rate": 0.00031172875030117676, + "loss": 0.0594, + "num_input_tokens_seen": 50794816, + "step": 39180 + }, + { + "epoch": 1.9146165685388317, + "grad_norm": 0.001033041742630303, + "learning_rate": 0.0003079446682686726, + "loss": 0.0971, + "num_input_tokens_seen": 50801184, + "step": 39185 + }, + { + "epoch": 1.914860870202526, + "grad_norm": 0.0013934376183897257, + "learning_rate": 0.0003041836706322465, + "loss": 0.0732, + "num_input_tokens_seen": 50807744, + "step": 39190 + }, + { + "epoch": 1.9151051718662204, + "grad_norm": 0.0011582054430618882, + "learning_rate": 0.0003004457579719011, + "loss": 0.0825, + "num_input_tokens_seen": 50813984, + "step": 39195 + }, + { + "epoch": 1.9153494735299148, + "grad_norm": 0.0011286742519587278, + "learning_rate": 0.00029673093086405867, + "loss": 0.0564, + "num_input_tokens_seen": 50820416, + "step": 39200 + }, + { + "epoch": 1.9153494735299148, + "eval_loss": 0.08342854678630829, + "eval_runtime": 402.4086, + "eval_samples_per_second": 90.418, + "eval_steps_per_second": 22.606, + "num_input_tokens_seen": 50820416, + "step": 39200 + }, + { + "epoch": 1.9155937751936092, + "grad_norm": 0.000654987059533596, + "learning_rate": 0.00029303918988159426, + "loss": 0.0636, + "num_input_tokens_seen": 50826816, + "step": 39205 + }, + { + "epoch": 1.9158380768573036, + "grad_norm": 0.0014937109081074595, + "learning_rate": 0.0002893705355938192, + "loss": 0.0617, + "num_input_tokens_seen": 50833376, + "step": 39210 + }, + { + "epoch": 1.9160823785209977, + "grad_norm": 0.0016937509644776583, + "learning_rate": 0.0002857249685664975, + "loss": 0.0813, + "num_input_tokens_seen": 50839936, + "step": 39215 + }, + { + "epoch": 1.916326680184692, + "grad_norm": 0.0019229995086789131, + "learning_rate": 0.0002821024893618129, + "loss": 0.0916, + "num_input_tokens_seen": 50846208, + "step": 39220 + }, + { + "epoch": 1.9165709818483863, + "grad_norm": 0.0014812445733696222, + "learning_rate": 0.0002785030985383852, + "loss": 0.0602, + "num_input_tokens_seen": 50852864, + "step": 39225 + }, + { + "epoch": 1.9168152835120806, + "grad_norm": 0.0019394431728869677, + "learning_rate": 0.00027492679665130356, + "loss": 0.0813, + "num_input_tokens_seen": 50859232, + "step": 39230 + }, + { + "epoch": 1.917059585175775, + "grad_norm": 0.0020534491632133722, + "learning_rate": 0.000271373584252077, + "loss": 0.0945, + "num_input_tokens_seen": 50865856, + "step": 39235 + }, + { + "epoch": 1.9173038868394694, + "grad_norm": 0.0015977972652763128, + "learning_rate": 0.00026784346188865046, + "loss": 0.084, + "num_input_tokens_seen": 50872192, + "step": 39240 + }, + { + "epoch": 1.9175481885031638, + "grad_norm": 0.0009631086722947657, + "learning_rate": 0.0002643364301054218, + "loss": 0.1142, + "num_input_tokens_seen": 50878752, + "step": 39245 + }, + { + "epoch": 1.9177924901668582, + "grad_norm": 0.0020411775913089514, + "learning_rate": 0.0002608524894431918, + "loss": 0.1083, + "num_input_tokens_seen": 50885280, + "step": 39250 + }, + { + "epoch": 1.9180367918305523, + "grad_norm": 0.0013910261914134026, + "learning_rate": 0.000257391640439264, + "loss": 0.1064, + "num_input_tokens_seen": 50891616, + "step": 39255 + }, + { + "epoch": 1.9182810934942467, + "grad_norm": 0.0009823169093579054, + "learning_rate": 0.00025395388362732806, + "loss": 0.0813, + "num_input_tokens_seen": 50898048, + "step": 39260 + }, + { + "epoch": 1.918525395157941, + "grad_norm": 0.0009151108679361641, + "learning_rate": 0.00025053921953751, + "loss": 0.0716, + "num_input_tokens_seen": 50904768, + "step": 39265 + }, + { + "epoch": 1.9187696968216352, + "grad_norm": 0.0008314131409861147, + "learning_rate": 0.00024714764869643855, + "loss": 0.0745, + "num_input_tokens_seen": 50911360, + "step": 39270 + }, + { + "epoch": 1.9190139984853296, + "grad_norm": 0.0010575789492577314, + "learning_rate": 0.0002437791716270954, + "loss": 0.096, + "num_input_tokens_seen": 50917856, + "step": 39275 + }, + { + "epoch": 1.919258300149024, + "grad_norm": 0.001997529761865735, + "learning_rate": 0.00024043378884896493, + "loss": 0.0961, + "num_input_tokens_seen": 50924416, + "step": 39280 + }, + { + "epoch": 1.9195026018127184, + "grad_norm": 0.0007777489372529089, + "learning_rate": 0.00023711150087793453, + "loss": 0.0962, + "num_input_tokens_seen": 50930560, + "step": 39285 + }, + { + "epoch": 1.9197469034764127, + "grad_norm": 0.001484760781750083, + "learning_rate": 0.000233812308226361, + "loss": 0.064, + "num_input_tokens_seen": 50936928, + "step": 39290 + }, + { + "epoch": 1.9199912051401071, + "grad_norm": 0.0011313359718769789, + "learning_rate": 0.00023053621140300406, + "loss": 0.0729, + "num_input_tokens_seen": 50943392, + "step": 39295 + }, + { + "epoch": 1.9202355068038013, + "grad_norm": 0.002348833018913865, + "learning_rate": 0.00022728321091307623, + "loss": 0.0921, + "num_input_tokens_seen": 50950240, + "step": 39300 + }, + { + "epoch": 1.9204798084674957, + "grad_norm": 0.0017651929520070553, + "learning_rate": 0.0002240533072582429, + "loss": 0.0684, + "num_input_tokens_seen": 50956960, + "step": 39305 + }, + { + "epoch": 1.92072411013119, + "grad_norm": 0.001665415707975626, + "learning_rate": 0.00022084650093658897, + "loss": 0.0926, + "num_input_tokens_seen": 50963648, + "step": 39310 + }, + { + "epoch": 1.9209684117948842, + "grad_norm": 0.001374374027363956, + "learning_rate": 0.0002176627924426522, + "loss": 0.1018, + "num_input_tokens_seen": 50969824, + "step": 39315 + }, + { + "epoch": 1.9212127134585786, + "grad_norm": 0.0031889791134744883, + "learning_rate": 0.0002145021822673898, + "loss": 0.082, + "num_input_tokens_seen": 50976448, + "step": 39320 + }, + { + "epoch": 1.921457015122273, + "grad_norm": 0.0014307685196399689, + "learning_rate": 0.00021136467089822862, + "loss": 0.0889, + "num_input_tokens_seen": 50982784, + "step": 39325 + }, + { + "epoch": 1.9217013167859673, + "grad_norm": 0.0013180478708818555, + "learning_rate": 0.00020825025881898162, + "loss": 0.0806, + "num_input_tokens_seen": 50989088, + "step": 39330 + }, + { + "epoch": 1.9219456184496617, + "grad_norm": 0.0015981163596734405, + "learning_rate": 0.0002051589465099479, + "loss": 0.0936, + "num_input_tokens_seen": 50995744, + "step": 39335 + }, + { + "epoch": 1.922189920113356, + "grad_norm": 0.001144491252489388, + "learning_rate": 0.0002020907344478462, + "loss": 0.0682, + "num_input_tokens_seen": 51001984, + "step": 39340 + }, + { + "epoch": 1.9224342217770503, + "grad_norm": 0.0010383836925029755, + "learning_rate": 0.0001990456231058313, + "loss": 0.0863, + "num_input_tokens_seen": 51008416, + "step": 39345 + }, + { + "epoch": 1.9226785234407446, + "grad_norm": 0.001550961984321475, + "learning_rate": 0.00019602361295349423, + "loss": 0.0643, + "num_input_tokens_seen": 51014880, + "step": 39350 + }, + { + "epoch": 1.922922825104439, + "grad_norm": 0.0013245169539004564, + "learning_rate": 0.0001930247044568789, + "loss": 0.0593, + "num_input_tokens_seen": 51021504, + "step": 39355 + }, + { + "epoch": 1.9231671267681332, + "grad_norm": 0.0010162844555452466, + "learning_rate": 0.00019004889807843205, + "loss": 0.0554, + "num_input_tokens_seen": 51028736, + "step": 39360 + }, + { + "epoch": 1.9234114284318276, + "grad_norm": 0.0010711103677749634, + "learning_rate": 0.00018709619427708656, + "loss": 0.0885, + "num_input_tokens_seen": 51034944, + "step": 39365 + }, + { + "epoch": 1.923655730095522, + "grad_norm": 0.001125076785683632, + "learning_rate": 0.00018416659350817822, + "loss": 0.0626, + "num_input_tokens_seen": 51041376, + "step": 39370 + }, + { + "epoch": 1.9239000317592163, + "grad_norm": 0.0007801977335475385, + "learning_rate": 0.00018126009622346229, + "loss": 0.0807, + "num_input_tokens_seen": 51047488, + "step": 39375 + }, + { + "epoch": 1.9241443334229107, + "grad_norm": 0.0012713914038613439, + "learning_rate": 0.00017837670287119687, + "loss": 0.0914, + "num_input_tokens_seen": 51054112, + "step": 39380 + }, + { + "epoch": 1.924388635086605, + "grad_norm": 0.0022532385773956776, + "learning_rate": 0.00017551641389602633, + "loss": 0.0835, + "num_input_tokens_seen": 51060544, + "step": 39385 + }, + { + "epoch": 1.9246329367502992, + "grad_norm": 0.0016893661813810468, + "learning_rate": 0.00017267922973903115, + "loss": 0.0805, + "num_input_tokens_seen": 51067008, + "step": 39390 + }, + { + "epoch": 1.9248772384139936, + "grad_norm": 0.0009278386132791638, + "learning_rate": 0.00016986515083774467, + "loss": 0.0707, + "num_input_tokens_seen": 51073696, + "step": 39395 + }, + { + "epoch": 1.9251215400776878, + "grad_norm": 0.0012422115541994572, + "learning_rate": 0.00016707417762611975, + "loss": 0.0749, + "num_input_tokens_seen": 51080832, + "step": 39400 + }, + { + "epoch": 1.9251215400776878, + "eval_loss": 0.083449587225914, + "eval_runtime": 402.8786, + "eval_samples_per_second": 90.313, + "eval_steps_per_second": 22.58, + "num_input_tokens_seen": 51080832, + "step": 39400 + }, + { + "epoch": 1.9253658417413821, + "grad_norm": 0.0016627043951302767, + "learning_rate": 0.00016430631053459543, + "loss": 0.0787, + "num_input_tokens_seen": 51087200, + "step": 39405 + }, + { + "epoch": 1.9256101434050765, + "grad_norm": 0.0008166012703441083, + "learning_rate": 0.0001615615499899803, + "loss": 0.0587, + "num_input_tokens_seen": 51093600, + "step": 39410 + }, + { + "epoch": 1.925854445068771, + "grad_norm": 0.0015342639526352286, + "learning_rate": 0.00015883989641556905, + "loss": 0.0755, + "num_input_tokens_seen": 51099936, + "step": 39415 + }, + { + "epoch": 1.9260987467324653, + "grad_norm": 0.0018844357691705227, + "learning_rate": 0.00015614135023105934, + "loss": 0.0738, + "num_input_tokens_seen": 51106336, + "step": 39420 + }, + { + "epoch": 1.9263430483961597, + "grad_norm": 0.0012609801487997174, + "learning_rate": 0.00015346591185261827, + "loss": 0.0869, + "num_input_tokens_seen": 51112864, + "step": 39425 + }, + { + "epoch": 1.926587350059854, + "grad_norm": 0.0011741038179025054, + "learning_rate": 0.00015081358169281576, + "loss": 0.083, + "num_input_tokens_seen": 51119552, + "step": 39430 + }, + { + "epoch": 1.9268316517235482, + "grad_norm": 0.001105376984924078, + "learning_rate": 0.00014818436016069135, + "loss": 0.0721, + "num_input_tokens_seen": 51125696, + "step": 39435 + }, + { + "epoch": 1.9270759533872426, + "grad_norm": 0.0017834455939009786, + "learning_rate": 0.00014557824766168735, + "loss": 0.0897, + "num_input_tokens_seen": 51132192, + "step": 39440 + }, + { + "epoch": 1.9273202550509367, + "grad_norm": 0.0013966833939775825, + "learning_rate": 0.00014299524459769896, + "loss": 0.0683, + "num_input_tokens_seen": 51138880, + "step": 39445 + }, + { + "epoch": 1.9275645567146311, + "grad_norm": 0.002196571324020624, + "learning_rate": 0.0001404353513670742, + "loss": 0.1051, + "num_input_tokens_seen": 51145312, + "step": 39450 + }, + { + "epoch": 1.9278088583783255, + "grad_norm": 0.001315773231908679, + "learning_rate": 0.0001378985683645806, + "loss": 0.0995, + "num_input_tokens_seen": 51151808, + "step": 39455 + }, + { + "epoch": 1.9280531600420199, + "grad_norm": 0.0014299562899395823, + "learning_rate": 0.0001353848959813886, + "loss": 0.0815, + "num_input_tokens_seen": 51158592, + "step": 39460 + }, + { + "epoch": 1.9282974617057143, + "grad_norm": 0.0010852640261873603, + "learning_rate": 0.00013289433460517142, + "loss": 0.0722, + "num_input_tokens_seen": 51165024, + "step": 39465 + }, + { + "epoch": 1.9285417633694086, + "grad_norm": 0.0016910441918298602, + "learning_rate": 0.00013042688462000518, + "loss": 0.0695, + "num_input_tokens_seen": 51171680, + "step": 39470 + }, + { + "epoch": 1.928786065033103, + "grad_norm": 0.0014493614435195923, + "learning_rate": 0.0001279825464063855, + "loss": 0.0858, + "num_input_tokens_seen": 51178176, + "step": 39475 + }, + { + "epoch": 1.9290303666967972, + "grad_norm": 0.0013048984110355377, + "learning_rate": 0.00012556132034126087, + "loss": 0.0635, + "num_input_tokens_seen": 51184736, + "step": 39480 + }, + { + "epoch": 1.9292746683604916, + "grad_norm": 0.001116492203436792, + "learning_rate": 0.0001231632067980326, + "loss": 0.0641, + "num_input_tokens_seen": 51191328, + "step": 39485 + }, + { + "epoch": 1.9295189700241857, + "grad_norm": 0.0012928876094520092, + "learning_rate": 0.00012078820614650486, + "loss": 0.0923, + "num_input_tokens_seen": 51197792, + "step": 39490 + }, + { + "epoch": 1.92976327168788, + "grad_norm": 0.002004611771553755, + "learning_rate": 0.00011843631875291804, + "loss": 0.1039, + "num_input_tokens_seen": 51204416, + "step": 39495 + }, + { + "epoch": 1.9300075733515745, + "grad_norm": 0.0017879531951621175, + "learning_rate": 0.00011610754497999863, + "loss": 0.1051, + "num_input_tokens_seen": 51210976, + "step": 39500 + }, + { + "epoch": 1.9302518750152688, + "grad_norm": 0.001581227290444076, + "learning_rate": 0.0001138018851868594, + "loss": 0.0867, + "num_input_tokens_seen": 51217120, + "step": 39505 + }, + { + "epoch": 1.9304961766789632, + "grad_norm": 0.001719644875265658, + "learning_rate": 0.0001115193397290326, + "loss": 0.095, + "num_input_tokens_seen": 51223488, + "step": 39510 + }, + { + "epoch": 1.9307404783426576, + "grad_norm": 0.0011554931988939643, + "learning_rate": 0.00010925990895856996, + "loss": 0.0655, + "num_input_tokens_seen": 51229504, + "step": 39515 + }, + { + "epoch": 1.930984780006352, + "grad_norm": 0.0010104216635227203, + "learning_rate": 0.00010702359322385946, + "loss": 0.1092, + "num_input_tokens_seen": 51235904, + "step": 39520 + }, + { + "epoch": 1.9312290816700461, + "grad_norm": 0.0012640218483284116, + "learning_rate": 0.00010481039286977523, + "loss": 0.0749, + "num_input_tokens_seen": 51242528, + "step": 39525 + }, + { + "epoch": 1.9314733833337405, + "grad_norm": 0.0019138590432703495, + "learning_rate": 0.00010262030823764423, + "loss": 0.0732, + "num_input_tokens_seen": 51249536, + "step": 39530 + }, + { + "epoch": 1.9317176849974347, + "grad_norm": 0.0009054734837263823, + "learning_rate": 0.00010045333966517966, + "loss": 0.0743, + "num_input_tokens_seen": 51255936, + "step": 39535 + }, + { + "epoch": 1.931961986661129, + "grad_norm": 0.001323827775195241, + "learning_rate": 9.83094874865642e-05, + "loss": 0.0872, + "num_input_tokens_seen": 51262624, + "step": 39540 + }, + { + "epoch": 1.9322062883248234, + "grad_norm": 0.0015549021773040295, + "learning_rate": 9.618875203241672e-05, + "loss": 0.0598, + "num_input_tokens_seen": 51269568, + "step": 39545 + }, + { + "epoch": 1.9324505899885178, + "grad_norm": 0.0025764619931578636, + "learning_rate": 9.409113362977561e-05, + "loss": 0.0875, + "num_input_tokens_seen": 51275616, + "step": 39550 + }, + { + "epoch": 1.9326948916522122, + "grad_norm": 0.0013755166437476873, + "learning_rate": 9.20166326020988e-05, + "loss": 0.0558, + "num_input_tokens_seen": 51282080, + "step": 39555 + }, + { + "epoch": 1.9329391933159066, + "grad_norm": 0.001439266954548657, + "learning_rate": 8.996524926933035e-05, + "loss": 0.0857, + "num_input_tokens_seen": 51288928, + "step": 39560 + }, + { + "epoch": 1.933183494979601, + "grad_norm": 0.002084410982206464, + "learning_rate": 8.793698394781723e-05, + "loss": 0.0635, + "num_input_tokens_seen": 51295168, + "step": 39565 + }, + { + "epoch": 1.9334277966432951, + "grad_norm": 0.0017804753733798862, + "learning_rate": 8.593183695030926e-05, + "loss": 0.0905, + "num_input_tokens_seen": 51301568, + "step": 39570 + }, + { + "epoch": 1.9336720983069895, + "grad_norm": 0.0012075647246092558, + "learning_rate": 8.39498085860757e-05, + "loss": 0.0816, + "num_input_tokens_seen": 51308000, + "step": 39575 + }, + { + "epoch": 1.9339163999706837, + "grad_norm": 0.0012234010500833392, + "learning_rate": 8.199089916072211e-05, + "loss": 0.077, + "num_input_tokens_seen": 51314208, + "step": 39580 + }, + { + "epoch": 1.934160701634378, + "grad_norm": 0.0021315969061106443, + "learning_rate": 8.005510897637346e-05, + "loss": 0.1108, + "num_input_tokens_seen": 51320384, + "step": 39585 + }, + { + "epoch": 1.9344050032980724, + "grad_norm": 0.001375738182105124, + "learning_rate": 7.8142438331541e-05, + "loss": 0.089, + "num_input_tokens_seen": 51326816, + "step": 39590 + }, + { + "epoch": 1.9346493049617668, + "grad_norm": 0.002512831473723054, + "learning_rate": 7.625288752117209e-05, + "loss": 0.0605, + "num_input_tokens_seen": 51332960, + "step": 39595 + }, + { + "epoch": 1.9348936066254612, + "grad_norm": 0.0013186202850192785, + "learning_rate": 7.4386456836667e-05, + "loss": 0.0717, + "num_input_tokens_seen": 51339424, + "step": 39600 + }, + { + "epoch": 1.9348936066254612, + "eval_loss": 0.08348038792610168, + "eval_runtime": 402.6247, + "eval_samples_per_second": 90.37, + "eval_steps_per_second": 22.594, + "num_input_tokens_seen": 51339424, + "step": 39600 + }, + { + "epoch": 1.9351379082891556, + "grad_norm": 0.001671448117122054, + "learning_rate": 7.254314656586214e-05, + "loss": 0.0884, + "num_input_tokens_seen": 51346080, + "step": 39605 + }, + { + "epoch": 1.93538220995285, + "grad_norm": 0.0005686490912921727, + "learning_rate": 7.07229569929968e-05, + "loss": 0.0526, + "num_input_tokens_seen": 51352640, + "step": 39610 + }, + { + "epoch": 1.935626511616544, + "grad_norm": 0.0015200544148683548, + "learning_rate": 6.892588839879643e-05, + "loss": 0.084, + "num_input_tokens_seen": 51358912, + "step": 39615 + }, + { + "epoch": 1.9358708132802385, + "grad_norm": 0.0012694926699623466, + "learning_rate": 6.71519410603727e-05, + "loss": 0.0673, + "num_input_tokens_seen": 51365760, + "step": 39620 + }, + { + "epoch": 1.9361151149439326, + "grad_norm": 0.0030589813832193613, + "learning_rate": 6.540111525129011e-05, + "loss": 0.0899, + "num_input_tokens_seen": 51372128, + "step": 39625 + }, + { + "epoch": 1.936359416607627, + "grad_norm": 0.0015268202405422926, + "learning_rate": 6.367341124154934e-05, + "loss": 0.075, + "num_input_tokens_seen": 51378656, + "step": 39630 + }, + { + "epoch": 1.9366037182713214, + "grad_norm": 0.0012761978432536125, + "learning_rate": 6.19688292975873e-05, + "loss": 0.0618, + "num_input_tokens_seen": 51385120, + "step": 39635 + }, + { + "epoch": 1.9368480199350158, + "grad_norm": 0.0017650444060564041, + "learning_rate": 6.0287369682260336e-05, + "loss": 0.0842, + "num_input_tokens_seen": 51391328, + "step": 39640 + }, + { + "epoch": 1.9370923215987101, + "grad_norm": 0.0019893739372491837, + "learning_rate": 5.8629032654894384e-05, + "loss": 0.0932, + "num_input_tokens_seen": 51397472, + "step": 39645 + }, + { + "epoch": 1.9373366232624045, + "grad_norm": 0.0017901664832606912, + "learning_rate": 5.699381847120155e-05, + "loss": 0.092, + "num_input_tokens_seen": 51403296, + "step": 39650 + }, + { + "epoch": 1.937580924926099, + "grad_norm": 0.0013542602537199855, + "learning_rate": 5.5381727383380094e-05, + "loss": 0.0796, + "num_input_tokens_seen": 51409408, + "step": 39655 + }, + { + "epoch": 1.937825226589793, + "grad_norm": 0.0018144809873774648, + "learning_rate": 5.379275964001451e-05, + "loss": 0.09, + "num_input_tokens_seen": 51416480, + "step": 39660 + }, + { + "epoch": 1.9380695282534874, + "grad_norm": 0.0009109515813179314, + "learning_rate": 5.222691548614211e-05, + "loss": 0.0776, + "num_input_tokens_seen": 51423296, + "step": 39665 + }, + { + "epoch": 1.9383138299171816, + "grad_norm": 0.0020758043974637985, + "learning_rate": 5.068419516323641e-05, + "loss": 0.0788, + "num_input_tokens_seen": 51429760, + "step": 39670 + }, + { + "epoch": 1.938558131580876, + "grad_norm": 0.0011800708016380668, + "learning_rate": 4.91645989092071e-05, + "loss": 0.0802, + "num_input_tokens_seen": 51436160, + "step": 39675 + }, + { + "epoch": 1.9388024332445704, + "grad_norm": 0.0010916474275290966, + "learning_rate": 4.7668126958400056e-05, + "loss": 0.0674, + "num_input_tokens_seen": 51442848, + "step": 39680 + }, + { + "epoch": 1.9390467349082647, + "grad_norm": 0.0011597948614507914, + "learning_rate": 4.619477954159734e-05, + "loss": 0.078, + "num_input_tokens_seen": 51448896, + "step": 39685 + }, + { + "epoch": 1.9392910365719591, + "grad_norm": 0.0008699095342308283, + "learning_rate": 4.4744556885983884e-05, + "loss": 0.0595, + "num_input_tokens_seen": 51455648, + "step": 39690 + }, + { + "epoch": 1.9395353382356535, + "grad_norm": 0.0015254340833052993, + "learning_rate": 4.331745921523078e-05, + "loss": 0.0724, + "num_input_tokens_seen": 51462528, + "step": 39695 + }, + { + "epoch": 1.9397796398993479, + "grad_norm": 0.0008257442386820912, + "learning_rate": 4.191348674937867e-05, + "loss": 0.0511, + "num_input_tokens_seen": 51469248, + "step": 39700 + }, + { + "epoch": 1.940023941563042, + "grad_norm": 0.00133255566470325, + "learning_rate": 4.0532639704971006e-05, + "loss": 0.0597, + "num_input_tokens_seen": 51476384, + "step": 39705 + }, + { + "epoch": 1.9402682432267364, + "grad_norm": 0.0017362851649522781, + "learning_rate": 3.917491829493747e-05, + "loss": 0.0777, + "num_input_tokens_seen": 51482624, + "step": 39710 + }, + { + "epoch": 1.9405125448904306, + "grad_norm": 0.0015232813311740756, + "learning_rate": 3.78403227286439e-05, + "loss": 0.0992, + "num_input_tokens_seen": 51488928, + "step": 39715 + }, + { + "epoch": 1.940756846554125, + "grad_norm": 0.0008294421131722629, + "learning_rate": 3.652885321192567e-05, + "loss": 0.0844, + "num_input_tokens_seen": 51494944, + "step": 39720 + }, + { + "epoch": 1.9410011482178193, + "grad_norm": 0.000902404310181737, + "learning_rate": 3.524050994702099e-05, + "loss": 0.0837, + "num_input_tokens_seen": 51501408, + "step": 39725 + }, + { + "epoch": 1.9412454498815137, + "grad_norm": 0.002448145765811205, + "learning_rate": 3.3975293132604276e-05, + "loss": 0.0703, + "num_input_tokens_seen": 51507616, + "step": 39730 + }, + { + "epoch": 1.941489751545208, + "grad_norm": 0.0016972963931038976, + "learning_rate": 3.2733202963786125e-05, + "loss": 0.0808, + "num_input_tokens_seen": 51514144, + "step": 39735 + }, + { + "epoch": 1.9417340532089025, + "grad_norm": 0.0014549988554790616, + "learning_rate": 3.15142396321133e-05, + "loss": 0.0715, + "num_input_tokens_seen": 51520128, + "step": 39740 + }, + { + "epoch": 1.9419783548725968, + "grad_norm": 0.0011390501167625189, + "learning_rate": 3.0318403325552132e-05, + "loss": 0.0932, + "num_input_tokens_seen": 51526656, + "step": 39745 + }, + { + "epoch": 1.942222656536291, + "grad_norm": 0.001228598295710981, + "learning_rate": 2.914569422855506e-05, + "loss": 0.0693, + "num_input_tokens_seen": 51532992, + "step": 39750 + }, + { + "epoch": 1.9424669581999854, + "grad_norm": 0.0015607532113790512, + "learning_rate": 2.7996112521927462e-05, + "loss": 0.0789, + "num_input_tokens_seen": 51539456, + "step": 39755 + }, + { + "epoch": 1.9427112598636795, + "grad_norm": 0.0020313402637839317, + "learning_rate": 2.68696583829775e-05, + "loss": 0.1056, + "num_input_tokens_seen": 51545600, + "step": 39760 + }, + { + "epoch": 1.942955561527374, + "grad_norm": 0.001044652657583356, + "learning_rate": 2.576633198539957e-05, + "loss": 0.0891, + "num_input_tokens_seen": 51551808, + "step": 39765 + }, + { + "epoch": 1.9431998631910683, + "grad_norm": 0.001628507161512971, + "learning_rate": 2.46861334993409e-05, + "loss": 0.0754, + "num_input_tokens_seen": 51558208, + "step": 39770 + }, + { + "epoch": 1.9434441648547627, + "grad_norm": 0.001494752592407167, + "learning_rate": 2.3629063091384903e-05, + "loss": 0.074, + "num_input_tokens_seen": 51564832, + "step": 39775 + }, + { + "epoch": 1.943688466518457, + "grad_norm": 0.0019620517268776894, + "learning_rate": 2.2595120924567834e-05, + "loss": 0.0804, + "num_input_tokens_seen": 51570944, + "step": 39780 + }, + { + "epoch": 1.9439327681821514, + "grad_norm": 0.0016814080299809575, + "learning_rate": 2.158430715829551e-05, + "loss": 0.0822, + "num_input_tokens_seen": 51577248, + "step": 39785 + }, + { + "epoch": 1.9441770698458456, + "grad_norm": 0.0011088871397078037, + "learning_rate": 2.059662194849321e-05, + "loss": 0.0582, + "num_input_tokens_seen": 51584000, + "step": 39790 + }, + { + "epoch": 1.94442137150954, + "grad_norm": 0.0009847857290878892, + "learning_rate": 1.9632065447422463e-05, + "loss": 0.072, + "num_input_tokens_seen": 51590464, + "step": 39795 + }, + { + "epoch": 1.9446656731732344, + "grad_norm": 0.0015141044277697802, + "learning_rate": 1.8690637803880916e-05, + "loss": 0.0663, + "num_input_tokens_seen": 51597120, + "step": 39800 + }, + { + "epoch": 1.9446656731732344, + "eval_loss": 0.08352842926979065, + "eval_runtime": 402.5895, + "eval_samples_per_second": 90.377, + "eval_steps_per_second": 22.596, + "num_input_tokens_seen": 51597120, + "step": 39800 + }, + { + "epoch": 1.9449099748369285, + "grad_norm": 0.0016871544066816568, + "learning_rate": 1.7772339163019123e-05, + "loss": 0.094, + "num_input_tokens_seen": 51604032, + "step": 39805 + }, + { + "epoch": 1.945154276500623, + "grad_norm": 0.0008465906139463186, + "learning_rate": 1.6877169666457138e-05, + "loss": 0.0687, + "num_input_tokens_seen": 51610464, + "step": 39810 + }, + { + "epoch": 1.9453985781643173, + "grad_norm": 0.0019627318251878023, + "learning_rate": 1.6005129452234532e-05, + "loss": 0.0829, + "num_input_tokens_seen": 51616928, + "step": 39815 + }, + { + "epoch": 1.9456428798280117, + "grad_norm": 0.0017312443815171719, + "learning_rate": 1.5156218654843733e-05, + "loss": 0.0924, + "num_input_tokens_seen": 51623520, + "step": 39820 + }, + { + "epoch": 1.945887181491706, + "grad_norm": 0.0014705673092976213, + "learning_rate": 1.4330437405196683e-05, + "loss": 0.0877, + "num_input_tokens_seen": 51630240, + "step": 39825 + }, + { + "epoch": 1.9461314831554004, + "grad_norm": 0.0015679296338930726, + "learning_rate": 1.352778583062486e-05, + "loss": 0.1067, + "num_input_tokens_seen": 51637152, + "step": 39830 + }, + { + "epoch": 1.9463757848190946, + "grad_norm": 0.0014822818338871002, + "learning_rate": 1.2748264054929237e-05, + "loss": 0.0802, + "num_input_tokens_seen": 51643648, + "step": 39835 + }, + { + "epoch": 1.946620086482789, + "grad_norm": 0.0007796856807544827, + "learning_rate": 1.1991872198297004e-05, + "loss": 0.098, + "num_input_tokens_seen": 51650272, + "step": 39840 + }, + { + "epoch": 1.9468643881464833, + "grad_norm": 0.0018442997243255377, + "learning_rate": 1.1258610377384847e-05, + "loss": 0.1195, + "num_input_tokens_seen": 51657024, + "step": 39845 + }, + { + "epoch": 1.9471086898101775, + "grad_norm": 0.0012369079049676657, + "learning_rate": 1.0548478705268982e-05, + "loss": 0.0853, + "num_input_tokens_seen": 51663424, + "step": 39850 + }, + { + "epoch": 1.9473529914738719, + "grad_norm": 0.0012117817532271147, + "learning_rate": 9.86147729147846e-06, + "loss": 0.0634, + "num_input_tokens_seen": 51669824, + "step": 39855 + }, + { + "epoch": 1.9475972931375662, + "grad_norm": 0.0013257364043965936, + "learning_rate": 9.197606241928557e-06, + "loss": 0.0862, + "num_input_tokens_seen": 51676448, + "step": 39860 + }, + { + "epoch": 1.9478415948012606, + "grad_norm": 0.002439408330246806, + "learning_rate": 8.556865659004042e-06, + "loss": 0.0802, + "num_input_tokens_seen": 51683328, + "step": 39865 + }, + { + "epoch": 1.948085896464955, + "grad_norm": 0.00171981705352664, + "learning_rate": 7.939255641525867e-06, + "loss": 0.0697, + "num_input_tokens_seen": 51689248, + "step": 39870 + }, + { + "epoch": 1.9483301981286494, + "grad_norm": 0.0008337883045896888, + "learning_rate": 7.344776284751164e-06, + "loss": 0.1147, + "num_input_tokens_seen": 51695968, + "step": 39875 + }, + { + "epoch": 1.9485744997923435, + "grad_norm": 0.00150409119669348, + "learning_rate": 6.773427680323296e-06, + "loss": 0.055, + "num_input_tokens_seen": 51702816, + "step": 39880 + }, + { + "epoch": 1.948818801456038, + "grad_norm": 0.004190626088529825, + "learning_rate": 6.225209916355112e-06, + "loss": 0.0933, + "num_input_tokens_seen": 51709184, + "step": 39885 + }, + { + "epoch": 1.9490631031197323, + "grad_norm": 0.0020914103370159864, + "learning_rate": 5.7001230774123e-06, + "loss": 0.0884, + "num_input_tokens_seen": 51715904, + "step": 39890 + }, + { + "epoch": 1.9493074047834265, + "grad_norm": 0.0033902600407600403, + "learning_rate": 5.198167244446772e-06, + "loss": 0.1167, + "num_input_tokens_seen": 51722016, + "step": 39895 + }, + { + "epoch": 1.9495517064471208, + "grad_norm": 0.0011896920623257756, + "learning_rate": 4.71934249487993e-06, + "loss": 0.0694, + "num_input_tokens_seen": 51728128, + "step": 39900 + }, + { + "epoch": 1.9497960081108152, + "grad_norm": 0.0027725358959287405, + "learning_rate": 4.2636489025527075e-06, + "loss": 0.0759, + "num_input_tokens_seen": 51734368, + "step": 39905 + }, + { + "epoch": 1.9500403097745096, + "grad_norm": 0.0006452453671954572, + "learning_rate": 3.831086537742223e-06, + "loss": 0.0773, + "num_input_tokens_seen": 51740768, + "step": 39910 + }, + { + "epoch": 1.950284611438204, + "grad_norm": 0.004577306564897299, + "learning_rate": 3.4216554671451236e-06, + "loss": 0.1004, + "num_input_tokens_seen": 51748000, + "step": 39915 + }, + { + "epoch": 1.9505289131018984, + "grad_norm": 0.001094386330805719, + "learning_rate": 3.035355753894242e-06, + "loss": 0.078, + "num_input_tokens_seen": 51754432, + "step": 39920 + }, + { + "epoch": 1.9507732147655925, + "grad_norm": 0.0015341052785515785, + "learning_rate": 2.6721874575752477e-06, + "loss": 0.0724, + "num_input_tokens_seen": 51761120, + "step": 39925 + }, + { + "epoch": 1.951017516429287, + "grad_norm": 0.0013513618614524603, + "learning_rate": 2.3321506341933418e-06, + "loss": 0.0877, + "num_input_tokens_seen": 51767712, + "step": 39930 + }, + { + "epoch": 1.951261818092981, + "grad_norm": 0.0012079617008566856, + "learning_rate": 2.0152453361732546e-06, + "loss": 0.0735, + "num_input_tokens_seen": 51773824, + "step": 39935 + }, + { + "epoch": 1.9515061197566754, + "grad_norm": 0.00171079789288342, + "learning_rate": 1.7214716123925554e-06, + "loss": 0.0753, + "num_input_tokens_seen": 51780064, + "step": 39940 + }, + { + "epoch": 1.9517504214203698, + "grad_norm": 0.0016099673230201006, + "learning_rate": 1.4508295081649968e-06, + "loss": 0.0831, + "num_input_tokens_seen": 51786752, + "step": 39945 + }, + { + "epoch": 1.9519947230840642, + "grad_norm": 0.0016678298125043511, + "learning_rate": 1.2033190652238623e-06, + "loss": 0.0903, + "num_input_tokens_seen": 51793056, + "step": 39950 + }, + { + "epoch": 1.9522390247477586, + "grad_norm": 0.0027258889749646187, + "learning_rate": 9.78940321721966e-07, + "loss": 0.0906, + "num_input_tokens_seen": 51799552, + "step": 39955 + }, + { + "epoch": 1.952483326411453, + "grad_norm": 0.0010730818612501025, + "learning_rate": 7.776933122816132e-07, + "loss": 0.0659, + "num_input_tokens_seen": 51805824, + "step": 39960 + }, + { + "epoch": 1.9527276280751473, + "grad_norm": 0.0015369560569524765, + "learning_rate": 5.99578067927986e-07, + "loss": 0.0938, + "num_input_tokens_seen": 51812544, + "step": 39965 + }, + { + "epoch": 1.9529719297388415, + "grad_norm": 0.0013758064014837146, + "learning_rate": 4.445946161224512e-07, + "loss": 0.0606, + "num_input_tokens_seen": 51819520, + "step": 39970 + }, + { + "epoch": 1.9532162314025359, + "grad_norm": 0.0010877670720219612, + "learning_rate": 3.127429807792126e-07, + "loss": 0.0719, + "num_input_tokens_seen": 51826336, + "step": 39975 + }, + { + "epoch": 1.95346053306623, + "grad_norm": 0.0012753023765981197, + "learning_rate": 2.040231822320049e-07, + "loss": 0.081, + "num_input_tokens_seen": 51833216, + "step": 39980 + }, + { + "epoch": 1.9537048347299244, + "grad_norm": 0.0011382832890376449, + "learning_rate": 1.1843523723409354e-07, + "loss": 0.0891, + "num_input_tokens_seen": 51840096, + "step": 39985 + }, + { + "epoch": 1.9539491363936188, + "grad_norm": 0.0010573217878118157, + "learning_rate": 5.597915897492811e-08, + "loss": 0.0682, + "num_input_tokens_seen": 51846688, + "step": 39990 + }, + { + "epoch": 1.9541934380573132, + "grad_norm": 0.0013175917556509376, + "learning_rate": 1.6654957113448885e-08, + "loss": 0.0871, + "num_input_tokens_seen": 51852640, + "step": 39995 + }, + { + "epoch": 1.9544377397210075, + "grad_norm": 0.0011914661154150963, + "learning_rate": 4.626377114735902e-10, + "loss": 0.0783, + "num_input_tokens_seen": 51858816, + "step": 40000 + }, + { + "epoch": 1.9544377397210075, + "eval_loss": 0.08348309993743896, + "eval_runtime": 402.6122, + "eval_samples_per_second": 90.372, + "eval_steps_per_second": 22.595, + "num_input_tokens_seen": 51858816, + "step": 40000 + }, + { + "epoch": 1.9544377397210075, + "num_input_tokens_seen": 51858816, + "step": 40000, + "total_flos": 2.1715238142060134e+17, + "train_loss": 0.10561876927614212, + "train_runtime": 96179.8765, + "train_samples_per_second": 6.654, + "train_steps_per_second": 0.416 + } + ], + "logging_steps": 5, + "max_steps": 40000, + "num_input_tokens_seen": 51858816, + "num_train_epochs": 2, + "save_steps": 200, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 2.1715238142060134e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +}