{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.13661868748475237,
  "eval_steps": 10,
  "global_step": 560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00024396194193705782,
      "grad_norm": 4.489601135253906,
      "learning_rate": 2.4999420463141455e-07,
      "loss": 3.306,
      "step": 1
    },
    {
      "epoch": 0.00024396194193705782,
      "eval_loss": 2.9874014854431152,
      "eval_runtime": 85.0857,
      "eval_samples_per_second": 3.009,
      "eval_steps_per_second": 0.752,
      "step": 1
    },
    {
      "epoch": 0.00048792388387411563,
      "grad_norm": 4.510775566101074,
      "learning_rate": 2.4998840671678217e-07,
      "loss": 2.7398,
      "step": 2
    },
    {
      "epoch": 0.0007318858258111735,
      "grad_norm": 5.426989555358887,
      "learning_rate": 2.499826062544247e-07,
      "loss": 3.3804,
      "step": 3
    },
    {
      "epoch": 0.0009758477677482313,
      "grad_norm": 3.3621065616607666,
      "learning_rate": 2.4997680324266246e-07,
      "loss": 2.8696,
      "step": 4
    },
    {
      "epoch": 0.0012198097096852891,
      "grad_norm": 4.033163070678711,
      "learning_rate": 2.499709976798144e-07,
      "loss": 3.4674,
      "step": 5
    },
    {
      "epoch": 0.001463771651622347,
      "grad_norm": 3.398115634918213,
      "learning_rate": 2.4996518956419777e-07,
      "loss": 3.3783,
      "step": 6
    },
    {
      "epoch": 0.0017077335935594047,
      "grad_norm": 2.1340525150299072,
      "learning_rate": 2.499593788941286e-07,
      "loss": 2.6618,
      "step": 7
    },
    {
      "epoch": 0.0019516955354964625,
      "grad_norm": 2.8017404079437256,
      "learning_rate": 2.499535656679212e-07,
      "loss": 3.1798,
      "step": 8
    },
    {
      "epoch": 0.0021956574774335204,
      "grad_norm": 2.3321993350982666,
      "learning_rate": 2.499477498838886e-07,
      "loss": 3.1456,
      "step": 9
    },
    {
      "epoch": 0.0024396194193705783,
      "grad_norm": 1.805264949798584,
      "learning_rate": 2.4994193154034227e-07,
      "loss": 2.7338,
      "step": 10
    },
    {
      "epoch": 0.0024396194193705783,
      "eval_loss": 2.8181073665618896,
      "eval_runtime": 82.6944,
      "eval_samples_per_second": 3.096,
      "eval_steps_per_second": 0.774,
      "step": 10
    },
    {
      "epoch": 0.002683581361307636,
      "grad_norm": 2.295260190963745,
      "learning_rate": 2.499361106355922e-07,
      "loss": 2.876,
      "step": 11
    },
    {
      "epoch": 0.002927543303244694,
      "grad_norm": 1.9248020648956299,
      "learning_rate": 2.499302871679468e-07,
      "loss": 3.0328,
      "step": 12
    },
    {
      "epoch": 0.0031715052451817514,
      "grad_norm": 1.5429915189743042,
      "learning_rate": 2.4992446113571303e-07,
      "loss": 2.7445,
      "step": 13
    },
    {
      "epoch": 0.0034154671871188093,
      "grad_norm": 1.659693717956543,
      "learning_rate": 2.4991863253719657e-07,
      "loss": 2.9829,
      "step": 14
    },
    {
      "epoch": 0.003659429129055867,
      "grad_norm": 1.6439993381500244,
      "learning_rate": 2.4991280137070126e-07,
      "loss": 2.7864,
      "step": 15
    },
    {
      "epoch": 0.003903391070992925,
      "grad_norm": 1.731806993484497,
      "learning_rate": 2.499069676345297e-07,
      "loss": 2.9397,
      "step": 16
    },
    {
      "epoch": 0.004147353012929983,
      "grad_norm": 1.397567629814148,
      "learning_rate": 2.499011313269829e-07,
      "loss": 2.7251,
      "step": 17
    },
    {
      "epoch": 0.004391314954867041,
      "grad_norm": 1.5513560771942139,
      "learning_rate": 2.498952924463603e-07,
      "loss": 2.9325,
      "step": 18
    },
    {
      "epoch": 0.004635276896804099,
      "grad_norm": 1.7622836828231812,
      "learning_rate": 2.498894509909601e-07,
      "loss": 2.7478,
      "step": 19
    },
    {
      "epoch": 0.0048792388387411565,
      "grad_norm": 1.4812703132629395,
      "learning_rate": 2.4988360695907864e-07,
      "loss": 2.7757,
      "step": 20
    },
    {
      "epoch": 0.0048792388387411565,
      "eval_loss": 2.6998705863952637,
      "eval_runtime": 82.4721,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 20
    },
    {
      "epoch": 0.005123200780678214,
      "grad_norm": 1.8131576776504517,
      "learning_rate": 2.49877760349011e-07,
      "loss": 2.5977,
      "step": 21
    },
    {
      "epoch": 0.005367162722615272,
      "grad_norm": 1.084616780281067,
      "learning_rate": 2.498719111590508e-07,
      "loss": 2.3549,
      "step": 22
    },
    {
      "epoch": 0.00561112466455233,
      "grad_norm": 1.909567952156067,
      "learning_rate": 2.498660593874899e-07,
      "loss": 3.0033,
      "step": 23
    },
    {
      "epoch": 0.005855086606489388,
      "grad_norm": 1.3913578987121582,
      "learning_rate": 2.4986020503261886e-07,
      "loss": 2.6655,
      "step": 24
    },
    {
      "epoch": 0.006099048548426446,
      "grad_norm": 1.499211311340332,
      "learning_rate": 2.498543480927266e-07,
      "loss": 2.7763,
      "step": 25
    },
    {
      "epoch": 0.006343010490363503,
      "grad_norm": 1.3160173892974854,
      "learning_rate": 2.4984848856610065e-07,
      "loss": 2.7146,
      "step": 26
    },
    {
      "epoch": 0.006586972432300561,
      "grad_norm": 1.4656383991241455,
      "learning_rate": 2.4984262645102706e-07,
      "loss": 2.8071,
      "step": 27
    },
    {
      "epoch": 0.006830934374237619,
      "grad_norm": 1.5149258375167847,
      "learning_rate": 2.4983676174579014e-07,
      "loss": 2.8613,
      "step": 28
    },
    {
      "epoch": 0.0070748963161746765,
      "grad_norm": 1.3552800416946411,
      "learning_rate": 2.498308944486729e-07,
      "loss": 2.6073,
      "step": 29
    },
    {
      "epoch": 0.007318858258111734,
      "grad_norm": 1.8789068460464478,
      "learning_rate": 2.4982502455795676e-07,
      "loss": 2.8036,
      "step": 30
    },
    {
      "epoch": 0.007318858258111734,
      "eval_loss": 2.5998635292053223,
      "eval_runtime": 82.4021,
      "eval_samples_per_second": 3.107,
      "eval_steps_per_second": 0.777,
      "step": 30
    },
    {
      "epoch": 0.007562820200048792,
      "grad_norm": 1.5838264226913452,
      "learning_rate": 2.498191520719216e-07,
      "loss": 2.7908,
      "step": 31
    },
    {
      "epoch": 0.00780678214198585,
      "grad_norm": 1.575810194015503,
      "learning_rate": 2.4981327698884575e-07,
      "loss": 2.5728,
      "step": 32
    },
    {
      "epoch": 0.008050744083922909,
      "grad_norm": 1.1017578840255737,
      "learning_rate": 2.498073993070061e-07,
      "loss": 2.519,
      "step": 33
    },
    {
      "epoch": 0.008294706025859966,
      "grad_norm": 1.5795230865478516,
      "learning_rate": 2.49801519024678e-07,
      "loss": 2.8713,
      "step": 34
    },
    {
      "epoch": 0.008538667967797023,
      "grad_norm": 1.3720916509628296,
      "learning_rate": 2.497956361401352e-07,
      "loss": 2.6911,
      "step": 35
    },
    {
      "epoch": 0.008782629909734082,
      "grad_norm": 1.3356428146362305,
      "learning_rate": 2.4978975065165004e-07,
      "loss": 2.5879,
      "step": 36
    },
    {
      "epoch": 0.009026591851671139,
      "grad_norm": 2.031726121902466,
      "learning_rate": 2.497838625574932e-07,
      "loss": 2.9549,
      "step": 37
    },
    {
      "epoch": 0.009270553793608197,
      "grad_norm": 1.4513427019119263,
      "learning_rate": 2.497779718559339e-07,
      "loss": 2.8033,
      "step": 38
    },
    {
      "epoch": 0.009514515735545254,
      "grad_norm": 1.4715417623519897,
      "learning_rate": 2.497720785452398e-07,
      "loss": 2.5233,
      "step": 39
    },
    {
      "epoch": 0.009758477677482313,
      "grad_norm": 1.3367327451705933,
      "learning_rate": 2.497661826236771e-07,
      "loss": 2.6558,
      "step": 40
    },
    {
      "epoch": 0.009758477677482313,
      "eval_loss": 2.5077946186065674,
      "eval_runtime": 82.473,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 40
    },
    {
      "epoch": 0.01000243961941937,
      "grad_norm": 1.8079203367233276,
      "learning_rate": 2.497602840895103e-07,
      "loss": 2.7062,
      "step": 41
    },
    {
      "epoch": 0.010246401561356429,
      "grad_norm": 1.297031283378601,
      "learning_rate": 2.4975438294100266e-07,
      "loss": 2.4938,
      "step": 42
    },
    {
      "epoch": 0.010490363503293486,
      "grad_norm": 2.0549025535583496,
      "learning_rate": 2.497484791764155e-07,
      "loss": 2.8457,
      "step": 43
    },
    {
      "epoch": 0.010734325445230545,
      "grad_norm": 2.118145227432251,
      "learning_rate": 2.4974257279400897e-07,
      "loss": 2.6677,
      "step": 44
    },
    {
      "epoch": 0.010978287387167602,
      "grad_norm": 1.6570909023284912,
      "learning_rate": 2.497366637920414e-07,
      "loss": 2.7371,
      "step": 45
    },
    {
      "epoch": 0.01122224932910466,
      "grad_norm": 2.100497007369995,
      "learning_rate": 2.497307521687697e-07,
      "loss": 2.7241,
      "step": 46
    },
    {
      "epoch": 0.011466211271041717,
      "grad_norm": 1.4280970096588135,
      "learning_rate": 2.497248379224492e-07,
      "loss": 2.53,
      "step": 47
    },
    {
      "epoch": 0.011710173212978776,
      "grad_norm": 1.6932001113891602,
      "learning_rate": 2.497189210513339e-07,
      "loss": 2.815,
      "step": 48
    },
    {
      "epoch": 0.011954135154915833,
      "grad_norm": 1.8314259052276611,
      "learning_rate": 2.497130015536758e-07,
      "loss": 2.7558,
      "step": 49
    },
    {
      "epoch": 0.012198097096852892,
      "grad_norm": 1.3970531225204468,
      "learning_rate": 2.497070794277257e-07,
      "loss": 2.5018,
      "step": 50
    },
    {
      "epoch": 0.012198097096852892,
      "eval_loss": 2.428684949874878,
      "eval_runtime": 82.4898,
      "eval_samples_per_second": 3.103,
      "eval_steps_per_second": 0.776,
      "step": 50
    },
    {
      "epoch": 0.012442059038789949,
      "grad_norm": 2.6512253284454346,
      "learning_rate": 2.497011546717327e-07,
      "loss": 2.6264,
      "step": 51
    },
    {
      "epoch": 0.012686020980727006,
      "grad_norm": 1.2588273286819458,
      "learning_rate": 2.496952272839445e-07,
      "loss": 2.3415,
      "step": 52
    },
    {
      "epoch": 0.012929982922664065,
      "grad_norm": 1.6063730716705322,
      "learning_rate": 2.4968929726260705e-07,
      "loss": 2.4857,
      "step": 53
    },
    {
      "epoch": 0.013173944864601122,
      "grad_norm": 1.344925045967102,
      "learning_rate": 2.4968336460596485e-07,
      "loss": 2.4742,
      "step": 54
    },
    {
      "epoch": 0.01341790680653818,
      "grad_norm": 1.141435146331787,
      "learning_rate": 2.4967742931226075e-07,
      "loss": 2.2668,
      "step": 55
    },
    {
      "epoch": 0.013661868748475237,
      "grad_norm": 1.4117755889892578,
      "learning_rate": 2.4967149137973625e-07,
      "loss": 2.4959,
      "step": 56
    },
    {
      "epoch": 0.013905830690412296,
      "grad_norm": 1.292641520500183,
      "learning_rate": 2.496655508066309e-07,
      "loss": 2.3535,
      "step": 57
    },
    {
      "epoch": 0.014149792632349353,
      "grad_norm": 1.3126784563064575,
      "learning_rate": 2.4965960759118313e-07,
      "loss": 2.3842,
      "step": 58
    },
    {
      "epoch": 0.014393754574286412,
      "grad_norm": 1.4474728107452393,
      "learning_rate": 2.4965366173162953e-07,
      "loss": 2.5879,
      "step": 59
    },
    {
      "epoch": 0.014637716516223469,
      "grad_norm": 1.4832170009613037,
      "learning_rate": 2.4964771322620516e-07,
      "loss": 2.5618,
      "step": 60
    },
    {
      "epoch": 0.014637716516223469,
      "eval_loss": 2.370858907699585,
      "eval_runtime": 82.5008,
      "eval_samples_per_second": 3.103,
      "eval_steps_per_second": 0.776,
      "step": 60
    },
    {
      "epoch": 0.014881678458160527,
      "grad_norm": 3.201457977294922,
      "learning_rate": 2.4964176207314356e-07,
      "loss": 2.4885,
      "step": 61
    },
    {
      "epoch": 0.015125640400097584,
      "grad_norm": 1.71616530418396,
      "learning_rate": 2.496358082706767e-07,
      "loss": 2.5725,
      "step": 62
    },
    {
      "epoch": 0.015369602342034643,
      "grad_norm": 1.8521833419799805,
      "learning_rate": 2.4962985181703483e-07,
      "loss": 2.4873,
      "step": 63
    },
    {
      "epoch": 0.0156135642839717,
      "grad_norm": 1.3216487169265747,
      "learning_rate": 2.496238927104469e-07,
      "loss": 2.4606,
      "step": 64
    },
    {
      "epoch": 0.015857526225908757,
      "grad_norm": 1.3042290210723877,
      "learning_rate": 2.4961793094913995e-07,
      "loss": 2.5374,
      "step": 65
    },
    {
      "epoch": 0.016101488167845818,
      "grad_norm": 1.4314343929290771,
      "learning_rate": 2.4961196653133975e-07,
      "loss": 2.5357,
      "step": 66
    },
    {
      "epoch": 0.016345450109782875,
      "grad_norm": 1.6834615468978882,
      "learning_rate": 2.4960599945527027e-07,
      "loss": 2.2217,
      "step": 67
    },
    {
      "epoch": 0.01658941205171993,
      "grad_norm": 1.1532320976257324,
      "learning_rate": 2.49600029719154e-07,
      "loss": 2.4338,
      "step": 68
    },
    {
      "epoch": 0.01683337399365699,
      "grad_norm": 1.009998083114624,
      "learning_rate": 2.495940573212118e-07,
      "loss": 2.18,
      "step": 69
    },
    {
      "epoch": 0.017077335935594046,
      "grad_norm": 1.4564158916473389,
      "learning_rate": 2.4958808225966306e-07,
      "loss": 2.3481,
      "step": 70
    },
    {
      "epoch": 0.017077335935594046,
      "eval_loss": 2.3252947330474854,
      "eval_runtime": 82.5205,
      "eval_samples_per_second": 3.102,
      "eval_steps_per_second": 0.776,
      "step": 70
    },
    {
      "epoch": 0.017321297877531106,
      "grad_norm": 1.5508018732070923,
      "learning_rate": 2.4958210453272533e-07,
      "loss": 2.5017,
      "step": 71
    },
    {
      "epoch": 0.017565259819468163,
      "grad_norm": 1.2118752002716064,
      "learning_rate": 2.4957612413861483e-07,
      "loss": 2.5198,
      "step": 72
    },
    {
      "epoch": 0.01780922176140522,
      "grad_norm": 1.1424647569656372,
      "learning_rate": 2.4957014107554603e-07,
      "loss": 2.4275,
      "step": 73
    },
    {
      "epoch": 0.018053183703342277,
      "grad_norm": 1.1591920852661133,
      "learning_rate": 2.4956415534173195e-07,
      "loss": 2.2197,
      "step": 74
    },
    {
      "epoch": 0.018297145645279338,
      "grad_norm": 1.1998584270477295,
      "learning_rate": 2.495581669353838e-07,
      "loss": 2.2739,
      "step": 75
    },
    {
      "epoch": 0.018541107587216395,
      "grad_norm": 1.0552688837051392,
      "learning_rate": 2.4955217585471147e-07,
      "loss": 2.4085,
      "step": 76
    },
    {
      "epoch": 0.01878506952915345,
      "grad_norm": 1.0630302429199219,
      "learning_rate": 2.495461820979229e-07,
      "loss": 2.3622,
      "step": 77
    },
    {
      "epoch": 0.01902903147109051,
      "grad_norm": 1.6193199157714844,
      "learning_rate": 2.4954018566322477e-07,
      "loss": 2.4233,
      "step": 78
    },
    {
      "epoch": 0.01927299341302757,
      "grad_norm": 1.0845212936401367,
      "learning_rate": 2.4953418654882195e-07,
      "loss": 2.2942,
      "step": 79
    },
    {
      "epoch": 0.019516955354964626,
      "grad_norm": 0.9362667202949524,
      "learning_rate": 2.495281847529178e-07,
      "loss": 2.3475,
      "step": 80
    },
    {
      "epoch": 0.019516955354964626,
      "eval_loss": 2.2899718284606934,
      "eval_runtime": 82.7023,
      "eval_samples_per_second": 3.095,
      "eval_steps_per_second": 0.774,
      "step": 80
    },
    {
      "epoch": 0.019760917296901683,
      "grad_norm": 1.1640156507492065,
      "learning_rate": 2.4952218027371403e-07,
      "loss": 2.4911,
      "step": 81
    },
    {
      "epoch": 0.02000487923883874,
      "grad_norm": 1.158489465713501,
      "learning_rate": 2.495161731094107e-07,
      "loss": 2.3697,
      "step": 82
    },
    {
      "epoch": 0.0202488411807758,
      "grad_norm": 1.056389570236206,
      "learning_rate": 2.4951016325820637e-07,
      "loss": 2.3726,
      "step": 83
    },
    {
      "epoch": 0.020492803122712858,
      "grad_norm": 1.1232126951217651,
      "learning_rate": 2.4950415071829794e-07,
      "loss": 2.3631,
      "step": 84
    },
    {
      "epoch": 0.020736765064649915,
      "grad_norm": 1.4430733919143677,
      "learning_rate": 2.4949813548788067e-07,
      "loss": 2.4389,
      "step": 85
    },
    {
      "epoch": 0.02098072700658697,
      "grad_norm": 1.4792566299438477,
      "learning_rate": 2.4949211756514816e-07,
      "loss": 2.5851,
      "step": 86
    },
    {
      "epoch": 0.02122468894852403,
      "grad_norm": 0.8782404661178589,
      "learning_rate": 2.494860969482926e-07,
      "loss": 2.3258,
      "step": 87
    },
    {
      "epoch": 0.02146865089046109,
      "grad_norm": 0.9481968879699707,
      "learning_rate": 2.4948007363550424e-07,
      "loss": 2.3977,
      "step": 88
    },
    {
      "epoch": 0.021712612832398146,
      "grad_norm": 1.0738717317581177,
      "learning_rate": 2.4947404762497197e-07,
      "loss": 2.0767,
      "step": 89
    },
    {
      "epoch": 0.021956574774335203,
      "grad_norm": 1.3180803060531616,
      "learning_rate": 2.49468018914883e-07,
      "loss": 2.4085,
      "step": 90
    },
    {
      "epoch": 0.021956574774335203,
      "eval_loss": 2.2610299587249756,
      "eval_runtime": 82.4227,
      "eval_samples_per_second": 3.106,
      "eval_steps_per_second": 0.776,
      "step": 90
    },
    {
      "epoch": 0.02220053671627226,
      "grad_norm": 0.9324449896812439,
      "learning_rate": 2.4946198750342283e-07,
      "loss": 2.3142,
      "step": 91
    },
    {
      "epoch": 0.02244449865820932,
      "grad_norm": 1.5807453393936157,
      "learning_rate": 2.4945595338877547e-07,
      "loss": 2.3756,
      "step": 92
    },
    {
      "epoch": 0.022688460600146378,
      "grad_norm": 1.279068112373352,
      "learning_rate": 2.494499165691231e-07,
      "loss": 2.2482,
      "step": 93
    },
    {
      "epoch": 0.022932422542083435,
      "grad_norm": 1.6906729936599731,
      "learning_rate": 2.4944387704264644e-07,
      "loss": 2.3038,
      "step": 94
    },
    {
      "epoch": 0.02317638448402049,
      "grad_norm": 1.2444514036178589,
      "learning_rate": 2.494378348075246e-07,
      "loss": 2.1946,
      "step": 95
    },
    {
      "epoch": 0.023420346425957552,
      "grad_norm": 0.9085439443588257,
      "learning_rate": 2.494317898619349e-07,
      "loss": 2.0829,
      "step": 96
    },
    {
      "epoch": 0.02366430836789461,
      "grad_norm": 1.0624847412109375,
      "learning_rate": 2.4942574220405314e-07,
      "loss": 2.2917,
      "step": 97
    },
    {
      "epoch": 0.023908270309831666,
      "grad_norm": 0.9223533868789673,
      "learning_rate": 2.4941969183205344e-07,
      "loss": 2.4056,
      "step": 98
    },
    {
      "epoch": 0.024152232251768723,
      "grad_norm": 4.390754699707031,
      "learning_rate": 2.494136387441083e-07,
      "loss": 2.4544,
      "step": 99
    },
    {
      "epoch": 0.024396194193705784,
      "grad_norm": 0.999297559261322,
      "learning_rate": 2.494075829383886e-07,
      "loss": 2.2129,
      "step": 100
    },
    {
      "epoch": 0.024396194193705784,
      "eval_loss": 2.2379603385925293,
      "eval_runtime": 82.4809,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 100
    },
    {
      "epoch": 0.02464015613564284,
      "grad_norm": 0.9902031421661377,
      "learning_rate": 2.494015244130635e-07,
      "loss": 2.0829,
      "step": 101
    },
    {
      "epoch": 0.024884118077579898,
      "grad_norm": 1.0847697257995605,
      "learning_rate": 2.493954631663007e-07,
      "loss": 2.2557,
      "step": 102
    },
    {
      "epoch": 0.025128080019516955,
      "grad_norm": 0.8790014982223511,
      "learning_rate": 2.493893991962659e-07,
      "loss": 2.2532,
      "step": 103
    },
    {
      "epoch": 0.02537204196145401,
      "grad_norm": 0.8715433478355408,
      "learning_rate": 2.493833325011235e-07,
      "loss": 2.348,
      "step": 104
    },
    {
      "epoch": 0.025616003903391072,
      "grad_norm": 0.9393193125724792,
      "learning_rate": 2.4937726307903606e-07,
      "loss": 2.3185,
      "step": 105
    },
    {
      "epoch": 0.02585996584532813,
      "grad_norm": 1.0732641220092773,
      "learning_rate": 2.493711909281646e-07,
      "loss": 2.3618,
      "step": 106
    },
    {
      "epoch": 0.026103927787265186,
      "grad_norm": 1.067499041557312,
      "learning_rate": 2.493651160466685e-07,
      "loss": 2.446,
      "step": 107
    },
    {
      "epoch": 0.026347889729202243,
      "grad_norm": 1.0830148458480835,
      "learning_rate": 2.493590384327053e-07,
      "loss": 2.4026,
      "step": 108
    },
    {
      "epoch": 0.026591851671139304,
      "grad_norm": 1.4239816665649414,
      "learning_rate": 2.49352958084431e-07,
      "loss": 2.2756,
      "step": 109
    },
    {
      "epoch": 0.02683581361307636,
      "grad_norm": 0.7910580635070801,
      "learning_rate": 2.49346875e-07,
      "loss": 2.1831,
      "step": 110
    },
    {
      "epoch": 0.02683581361307636,
      "eval_loss": 2.2186379432678223,
      "eval_runtime": 82.443,
      "eval_samples_per_second": 3.105,
      "eval_steps_per_second": 0.776,
      "step": 110
    },
    {
      "epoch": 0.027079775555013418,
      "grad_norm": 1.1051030158996582,
      "learning_rate": 2.49340789177565e-07,
      "loss": 2.2737,
      "step": 111
    },
    {
      "epoch": 0.027323737496950475,
      "grad_norm": 1.0497264862060547,
      "learning_rate": 2.4933470061527687e-07,
      "loss": 2.2902,
      "step": 112
    },
    {
      "epoch": 0.027567699438887535,
      "grad_norm": 0.87137770652771,
      "learning_rate": 2.493286093112851e-07,
      "loss": 2.1482,
      "step": 113
    },
    {
      "epoch": 0.027811661380824592,
      "grad_norm": 1.1766656637191772,
      "learning_rate": 2.493225152637374e-07,
      "loss": 2.353,
      "step": 114
    },
    {
      "epoch": 0.02805562332276165,
      "grad_norm": 0.9225324988365173,
      "learning_rate": 2.4931641847077963e-07,
      "loss": 2.1774,
      "step": 115
    },
    {
      "epoch": 0.028299585264698706,
      "grad_norm": 0.8458165526390076,
      "learning_rate": 2.493103189305562e-07,
      "loss": 2.112,
      "step": 116
    },
    {
      "epoch": 0.028543547206635766,
      "grad_norm": 0.975180983543396,
      "learning_rate": 2.493042166412099e-07,
      "loss": 2.331,
      "step": 117
    },
    {
      "epoch": 0.028787509148572824,
      "grad_norm": 0.879942774772644,
      "learning_rate": 2.492981116008816e-07,
      "loss": 2.2939,
      "step": 118
    },
    {
      "epoch": 0.02903147109050988,
      "grad_norm": 1.5170252323150635,
      "learning_rate": 2.492920038077106e-07,
      "loss": 2.435,
      "step": 119
    },
    {
      "epoch": 0.029275433032446938,
      "grad_norm": 1.0312517881393433,
      "learning_rate": 2.492858932598346e-07,
      "loss": 2.2375,
      "step": 120
    },
    {
      "epoch": 0.029275433032446938,
      "eval_loss": 2.201087474822998,
      "eval_runtime": 82.5159,
      "eval_samples_per_second": 3.102,
      "eval_steps_per_second": 0.776,
      "step": 120
    },
    {
      "epoch": 0.029519394974383995,
      "grad_norm": 0.999895453453064,
      "learning_rate": 2.4927977995538954e-07,
      "loss": 2.2366,
      "step": 121
    },
    {
      "epoch": 0.029763356916321055,
      "grad_norm": 0.75639408826828,
      "learning_rate": 2.4927366389250973e-07,
      "loss": 2.1457,
      "step": 122
    },
    {
      "epoch": 0.030007318858258112,
      "grad_norm": 1.004939079284668,
      "learning_rate": 2.4926754506932774e-07,
      "loss": 2.2746,
      "step": 123
    },
    {
      "epoch": 0.03025128080019517,
      "grad_norm": 0.9373717308044434,
      "learning_rate": 2.4926142348397453e-07,
      "loss": 2.2899,
      "step": 124
    },
    {
      "epoch": 0.030495242742132226,
      "grad_norm": 22.97978401184082,
      "learning_rate": 2.492552991345792e-07,
      "loss": 2.3833,
      "step": 125
    },
    {
      "epoch": 0.030739204684069286,
      "grad_norm": 1.2713708877563477,
      "learning_rate": 2.4924917201926936e-07,
      "loss": 2.3817,
      "step": 126
    },
    {
      "epoch": 0.030983166626006343,
      "grad_norm": 1.0373756885528564,
      "learning_rate": 2.492430421361708e-07,
      "loss": 2.0911,
      "step": 127
    },
    {
      "epoch": 0.0312271285679434,
      "grad_norm": 1.1061681509017944,
      "learning_rate": 2.4923690948340783e-07,
      "loss": 2.2405,
      "step": 128
    },
    {
      "epoch": 0.03147109050988046,
      "grad_norm": 0.8704845309257507,
      "learning_rate": 2.4923077405910264e-07,
      "loss": 2.3703,
      "step": 129
    },
    {
      "epoch": 0.031715052451817514,
      "grad_norm": 0.8377108573913574,
      "learning_rate": 2.4922463586137616e-07,
      "loss": 2.1756,
      "step": 130
    },
    {
      "epoch": 0.031715052451817514,
      "eval_loss": 2.1864898204803467,
      "eval_runtime": 82.5766,
      "eval_samples_per_second": 3.1,
      "eval_steps_per_second": 0.775,
      "step": 130
    },
    {
      "epoch": 0.03195901439375457,
      "grad_norm": 0.9043044447898865,
      "learning_rate": 2.4921849488834745e-07,
      "loss": 2.3177,
      "step": 131
    },
    {
      "epoch": 0.032202976335691635,
      "grad_norm": 0.9630913138389587,
      "learning_rate": 2.4921235113813376e-07,
      "loss": 2.3257,
      "step": 132
    },
    {
      "epoch": 0.03244693827762869,
      "grad_norm": 0.9598456025123596,
      "learning_rate": 2.492062046088508e-07,
      "loss": 2.1792,
      "step": 133
    },
    {
      "epoch": 0.03269090021956575,
      "grad_norm": 0.9462944865226746,
      "learning_rate": 2.4920005529861254e-07,
      "loss": 2.1268,
      "step": 134
    },
    {
      "epoch": 0.032934862161502806,
      "grad_norm": 0.9108706116676331,
      "learning_rate": 2.491939032055311e-07,
      "loss": 2.2696,
      "step": 135
    },
    {
      "epoch": 0.03317882410343986,
      "grad_norm": 1.834842324256897,
      "learning_rate": 2.491877483277171e-07,
      "loss": 2.3871,
      "step": 136
    },
    {
      "epoch": 0.03342278604537692,
      "grad_norm": 0.9213249087333679,
      "learning_rate": 2.4918159066327943e-07,
      "loss": 2.1749,
      "step": 137
    },
    {
      "epoch": 0.03366674798731398,
      "grad_norm": 0.8780300617218018,
      "learning_rate": 2.49175430210325e-07,
      "loss": 2.2413,
      "step": 138
    },
    {
      "epoch": 0.033910709929251034,
      "grad_norm": 118.5926284790039,
      "learning_rate": 2.491692669669594e-07,
      "loss": 2.2684,
      "step": 139
    },
    {
      "epoch": 0.03415467187118809,
      "grad_norm": 0.8691902160644531,
      "learning_rate": 2.4916310093128616e-07,
      "loss": 2.1863,
      "step": 140
    },
    {
      "epoch": 0.03415467187118809,
      "eval_loss": 2.174090623855591,
      "eval_runtime": 82.5703,
      "eval_samples_per_second": 3.1,
      "eval_steps_per_second": 0.775,
      "step": 140
    },
    {
      "epoch": 0.034398633813125155,
      "grad_norm": 0.8839126825332642,
      "learning_rate": 2.491569321014073e-07,
      "loss": 2.2502,
      "step": 141
    },
    {
      "epoch": 0.03464259575506221,
      "grad_norm": 1.0043957233428955,
      "learning_rate": 2.49150760475423e-07,
      "loss": 2.2624,
      "step": 142
    },
    {
      "epoch": 0.03488655769699927,
      "grad_norm": 0.7548794746398926,
      "learning_rate": 2.4914458605143187e-07,
      "loss": 2.2126,
      "step": 143
    },
    {
      "epoch": 0.035130519638936326,
      "grad_norm": 1.0142180919647217,
      "learning_rate": 2.491384088275306e-07,
      "loss": 2.1853,
      "step": 144
    },
    {
      "epoch": 0.03537448158087338,
      "grad_norm": 0.9584774971008301,
      "learning_rate": 2.491322288018143e-07,
      "loss": 2.0712,
      "step": 145
    },
    {
      "epoch": 0.03561844352281044,
      "grad_norm": 1.2019151449203491,
      "learning_rate": 2.4912604597237626e-07,
      "loss": 2.1924,
      "step": 146
    },
    {
      "epoch": 0.0358624054647475,
      "grad_norm": 0.8215965628623962,
      "learning_rate": 2.4911986033730807e-07,
      "loss": 2.1359,
      "step": 147
    },
    {
      "epoch": 0.036106367406684554,
      "grad_norm": 0.8709486722946167,
      "learning_rate": 2.491136718946997e-07,
      "loss": 2.4015,
      "step": 148
    },
    {
      "epoch": 0.03635032934862162,
      "grad_norm": 0.7629618644714355,
      "learning_rate": 2.4910748064263914e-07,
      "loss": 2.2006,
      "step": 149
    },
    {
      "epoch": 0.036594291290558675,
      "grad_norm": 1.1265887022018433,
      "learning_rate": 2.491012865792129e-07,
      "loss": 2.2315,
      "step": 150
    },
    {
      "epoch": 0.036594291290558675,
      "eval_loss": 2.162862539291382,
      "eval_runtime": 82.4917,
      "eval_samples_per_second": 3.103,
      "eval_steps_per_second": 0.776,
      "step": 150
    },
    {
      "epoch": 0.03683825323249573,
      "grad_norm": 0.8692772388458252,
      "learning_rate": 2.490950897025056e-07,
      "loss": 2.1676,
      "step": 151
    },
    {
      "epoch": 0.03708221517443279,
      "grad_norm": 0.865197479724884,
      "learning_rate": 2.4908889001060015e-07,
      "loss": 2.2783,
      "step": 152
    },
    {
      "epoch": 0.037326177116369846,
      "grad_norm": 0.922754168510437,
      "learning_rate": 2.490826875015777e-07,
      "loss": 2.3296,
      "step": 153
    },
    {
      "epoch": 0.0375701390583069,
      "grad_norm": 0.9516766667366028,
      "learning_rate": 2.490764821735178e-07,
      "loss": 2.3431,
      "step": 154
    },
    {
      "epoch": 0.03781410100024396,
      "grad_norm": 0.9996930360794067,
      "learning_rate": 2.4907027402449803e-07,
      "loss": 2.2365,
      "step": 155
    },
    {
      "epoch": 0.03805806294218102,
      "grad_norm": 0.7445939779281616,
      "learning_rate": 2.4906406305259434e-07,
      "loss": 2.1925,
      "step": 156
    },
    {
      "epoch": 0.038302024884118074,
      "grad_norm": 0.8426290154457092,
      "learning_rate": 2.4905784925588094e-07,
      "loss": 2.2349,
      "step": 157
    },
    {
      "epoch": 0.03854598682605514,
      "grad_norm": 1.0926883220672607,
      "learning_rate": 2.4905163263243023e-07,
      "loss": 2.2426,
      "step": 158
    },
    {
      "epoch": 0.038789948767992195,
      "grad_norm": 1.0484980344772339,
      "learning_rate": 2.4904541318031294e-07,
      "loss": 2.1593,
      "step": 159
    },
    {
      "epoch": 0.03903391070992925,
      "grad_norm": 0.8670967817306519,
      "learning_rate": 2.49039190897598e-07,
      "loss": 2.2036,
      "step": 160
    },
    {
      "epoch": 0.03903391070992925,
      "eval_loss": 2.152311086654663,
      "eval_runtime": 82.7226,
      "eval_samples_per_second": 3.095,
      "eval_steps_per_second": 0.774,
      "step": 160
    },
    {
      "epoch": 0.03927787265186631,
      "grad_norm": 0.8905733227729797,
      "learning_rate": 2.490329657823525e-07,
      "loss": 2.0864,
      "step": 161
    },
    {
      "epoch": 0.039521834593803366,
      "grad_norm": 0.8731813430786133,
      "learning_rate": 2.490267378326419e-07,
      "loss": 2.218,
      "step": 162
    },
    {
      "epoch": 0.03976579653574042,
      "grad_norm": 0.7908375263214111,
      "learning_rate": 2.490205070465299e-07,
      "loss": 2.1147,
      "step": 163
    },
    {
      "epoch": 0.04000975847767748,
      "grad_norm": 0.9735328555107117,
      "learning_rate": 2.4901427342207823e-07,
      "loss": 2.1494,
      "step": 164
    },
    {
      "epoch": 0.04025372041961454,
      "grad_norm": 0.7927246689796448,
      "learning_rate": 2.490080369573472e-07,
      "loss": 2.1533,
      "step": 165
    },
    {
      "epoch": 0.0404976823615516,
      "grad_norm": 0.8604568839073181,
      "learning_rate": 2.4900179765039496e-07,
      "loss": 2.1147,
      "step": 166
    },
    {
      "epoch": 0.04074164430348866,
      "grad_norm": 1.905282735824585,
      "learning_rate": 2.489955554992782e-07,
      "loss": 2.1228,
      "step": 167
    },
    {
      "epoch": 0.040985606245425715,
      "grad_norm": 0.8245707154273987,
      "learning_rate": 2.489893105020518e-07,
      "loss": 2.0648,
      "step": 168
    },
    {
      "epoch": 0.04122956818736277,
      "grad_norm": 0.9537479281425476,
      "learning_rate": 2.489830626567686e-07,
      "loss": 2.2827,
      "step": 169
    },
    {
      "epoch": 0.04147353012929983,
      "grad_norm": 0.8153314590454102,
      "learning_rate": 2.4897681196148e-07,
      "loss": 2.1892,
      "step": 170
    },
    {
      "epoch": 0.04147353012929983,
      "eval_loss": 2.14363694190979,
      "eval_runtime": 84.0031,
      "eval_samples_per_second": 3.048,
      "eval_steps_per_second": 0.762,
      "step": 170
    },
    {
      "epoch": 0.041717492071236886,
      "grad_norm": 0.9862874150276184,
      "learning_rate": 2.4897055841423537e-07,
      "loss": 2.1974,
      "step": 171
    },
    {
      "epoch": 0.04196145401317394,
      "grad_norm": 0.7293988466262817,
      "learning_rate": 2.489643020130825e-07,
      "loss": 2.1471,
      "step": 172
    },
    {
      "epoch": 0.042205415955111,
      "grad_norm": 0.981706976890564,
      "learning_rate": 2.4895804275606724e-07,
      "loss": 2.1939,
      "step": 173
    },
    {
      "epoch": 0.04244937789704806,
      "grad_norm": 0.7976970076560974,
      "learning_rate": 2.489517806412337e-07,
      "loss": 2.3129,
      "step": 174
    },
    {
      "epoch": 0.04269333983898512,
      "grad_norm": 1.0813937187194824,
      "learning_rate": 2.4894551566662435e-07,
      "loss": 2.4131,
      "step": 175
    },
    {
      "epoch": 0.04293730178092218,
      "grad_norm": 0.7610160708427429,
      "learning_rate": 2.4893924783027967e-07,
      "loss": 2.2202,
      "step": 176
    },
    {
      "epoch": 0.043181263722859235,
      "grad_norm": 0.8845246434211731,
      "learning_rate": 2.4893297713023835e-07,
      "loss": 2.0994,
      "step": 177
    },
    {
      "epoch": 0.04342522566479629,
      "grad_norm": 0.8146999478340149,
      "learning_rate": 2.4892670356453745e-07,
      "loss": 2.2112,
      "step": 178
    },
    {
      "epoch": 0.04366918760673335,
      "grad_norm": 53.56632995605469,
      "learning_rate": 2.4892042713121207e-07,
      "loss": 2.2625,
      "step": 179
    },
    {
      "epoch": 0.043913149548670406,
      "grad_norm": 0.8157463073730469,
      "learning_rate": 2.4891414782829566e-07,
      "loss": 2.206,
      "step": 180
    },
    {
      "epoch": 0.043913149548670406,
      "eval_loss": 2.1358835697174072,
      "eval_runtime": 84.0237,
      "eval_samples_per_second": 3.047,
      "eval_steps_per_second": 0.762,
      "step": 180
    },
    {
      "epoch": 0.04415711149060746,
      "grad_norm": 0.776892364025116,
      "learning_rate": 2.4890786565381976e-07,
      "loss": 2.1461,
      "step": 181
    },
    {
      "epoch": 0.04440107343254452,
      "grad_norm": 1.0113416910171509,
      "learning_rate": 2.489015806058142e-07,
      "loss": 2.2795,
      "step": 182
    },
    {
      "epoch": 0.044645035374481584,
      "grad_norm": 1.4400883913040161,
      "learning_rate": 2.4889529268230683e-07,
      "loss": 2.252,
      "step": 183
    },
    {
      "epoch": 0.04488899731641864,
      "grad_norm": 1.5715802907943726,
      "learning_rate": 2.4888900188132405e-07,
      "loss": 2.254,
      "step": 184
    },
    {
      "epoch": 0.0451329592583557,
      "grad_norm": 0.8854827284812927,
      "learning_rate": 2.4888270820089003e-07,
      "loss": 2.2534,
      "step": 185
    },
    {
      "epoch": 0.045376921200292755,
      "grad_norm": 0.8536761403083801,
      "learning_rate": 2.488764116390274e-07,
      "loss": 2.243,
      "step": 186
    },
    {
      "epoch": 0.04562088314222981,
      "grad_norm": 1.156900405883789,
      "learning_rate": 2.488701121937568e-07,
      "loss": 2.2924,
      "step": 187
    },
    {
      "epoch": 0.04586484508416687,
      "grad_norm": 0.9670676589012146,
      "learning_rate": 2.488638098630973e-07,
      "loss": 2.1767,
      "step": 188
    },
    {
      "epoch": 0.046108807026103926,
      "grad_norm": 0.6777328848838806,
      "learning_rate": 2.4885750464506606e-07,
      "loss": 2.2506,
      "step": 189
    },
    {
      "epoch": 0.04635276896804098,
      "grad_norm": 1.1340001821517944,
      "learning_rate": 2.488511965376782e-07,
      "loss": 2.2462,
      "step": 190
    },
    {
      "epoch": 0.04635276896804098,
      "eval_loss": 2.1293463706970215,
      "eval_runtime": 82.7369,
      "eval_samples_per_second": 3.094,
      "eval_steps_per_second": 0.774,
      "step": 190
    },
    {
      "epoch": 0.04659673090997804,
      "grad_norm": 0.9202519655227661,
      "learning_rate": 2.488448855389473e-07,
      "loss": 2.264,
      "step": 191
    },
    {
      "epoch": 0.046840692851915104,
      "grad_norm": 0.7520004510879517,
      "learning_rate": 2.48838571646885e-07,
      "loss": 2.184,
      "step": 192
    },
    {
      "epoch": 0.04708465479385216,
      "grad_norm": 0.9756651520729065,
      "learning_rate": 2.488322548595012e-07,
      "loss": 2.2218,
      "step": 193
    },
    {
      "epoch": 0.04732861673578922,
      "grad_norm": 1.1608117818832397,
      "learning_rate": 2.488259351748038e-07,
      "loss": 2.3409,
      "step": 194
    },
    {
      "epoch": 0.047572578677726275,
      "grad_norm": 0.9417536854743958,
      "learning_rate": 2.48819612590799e-07,
      "loss": 2.2483,
      "step": 195
    },
    {
      "epoch": 0.04781654061966333,
      "grad_norm": 0.8184861540794373,
      "learning_rate": 2.4881328710549126e-07,
      "loss": 2.1357,
      "step": 196
    },
    {
      "epoch": 0.04806050256160039,
      "grad_norm": 0.8574642539024353,
      "learning_rate": 2.48806958716883e-07,
      "loss": 2.2997,
      "step": 197
    },
    {
      "epoch": 0.048304464503537446,
      "grad_norm": 0.762095034122467,
      "learning_rate": 2.488006274229749e-07,
      "loss": 2.1704,
      "step": 198
    },
    {
      "epoch": 0.0485484264454745,
      "grad_norm": 0.876278281211853,
      "learning_rate": 2.4879429322176583e-07,
      "loss": 2.1739,
      "step": 199
    },
    {
      "epoch": 0.04879238838741157,
      "grad_norm": 0.7141769528388977,
      "learning_rate": 2.4878795611125284e-07,
      "loss": 2.1301,
      "step": 200
    },
    {
      "epoch": 0.04879238838741157,
      "eval_loss": 2.1227312088012695,
      "eval_runtime": 82.5491,
      "eval_samples_per_second": 3.101,
      "eval_steps_per_second": 0.775,
      "step": 200
    },
    {
      "epoch": 0.049036350329348624,
      "grad_norm": 0.793533444404602,
      "learning_rate": 2.487816160894311e-07,
      "loss": 2.105,
      "step": 201
    },
    {
      "epoch": 0.04928031227128568,
      "grad_norm": 1.4064488410949707,
      "learning_rate": 2.4877527315429387e-07,
      "loss": 2.2669,
      "step": 202
    },
    {
      "epoch": 0.04952427421322274,
      "grad_norm": 0.7336916923522949,
      "learning_rate": 2.4876892730383267e-07,
      "loss": 2.2367,
      "step": 203
    },
    {
      "epoch": 0.049768236155159795,
      "grad_norm": 5.0201287269592285,
      "learning_rate": 2.4876257853603717e-07,
      "loss": 2.3826,
      "step": 204
    },
    {
      "epoch": 0.05001219809709685,
      "grad_norm": 1.0578281879425049,
      "learning_rate": 2.4875622684889513e-07,
      "loss": 2.178,
      "step": 205
    },
    {
      "epoch": 0.05025616003903391,
      "grad_norm": 0.7075783610343933,
      "learning_rate": 2.4874987224039246e-07,
      "loss": 2.2244,
      "step": 206
    },
    {
      "epoch": 0.050500121980970966,
      "grad_norm": 1.0379695892333984,
      "learning_rate": 2.4874351470851334e-07,
      "loss": 2.2114,
      "step": 207
    },
    {
      "epoch": 0.05074408392290802,
      "grad_norm": 0.8745626211166382,
      "learning_rate": 2.4873715425123986e-07,
      "loss": 2.2984,
      "step": 208
    },
    {
      "epoch": 0.05098804586484509,
      "grad_norm": 0.7211839556694031,
      "learning_rate": 2.4873079086655244e-07,
      "loss": 2.2514,
      "step": 209
    },
    {
      "epoch": 0.051232007806782144,
      "grad_norm": 1.0270675420761108,
      "learning_rate": 2.487244245524296e-07,
      "loss": 2.0273,
      "step": 210
    },
    {
      "epoch": 0.051232007806782144,
      "eval_loss": 2.1163525581359863,
      "eval_runtime": 82.5285,
      "eval_samples_per_second": 3.102,
      "eval_steps_per_second": 0.775,
      "step": 210
    },
    {
      "epoch": 0.0514759697487192,
      "grad_norm": 0.7041739225387573,
      "learning_rate": 2.487180553068481e-07,
      "loss": 2.2021,
      "step": 211
    },
    {
      "epoch": 0.05171993169065626,
      "grad_norm": 0.7967678904533386,
      "learning_rate": 2.487116831277826e-07,
      "loss": 2.0156,
      "step": 212
    },
    {
      "epoch": 0.051963893632593315,
      "grad_norm": 0.9619562029838562,
      "learning_rate": 2.4870530801320607e-07,
      "loss": 2.0662,
      "step": 213
    },
    {
      "epoch": 0.05220785557453037,
      "grad_norm": 0.6969118714332581,
      "learning_rate": 2.486989299610895e-07,
      "loss": 2.1831,
      "step": 214
    },
    {
      "epoch": 0.05245181751646743,
      "grad_norm": 0.9280874729156494,
      "learning_rate": 2.4869254896940207e-07,
      "loss": 2.1238,
      "step": 215
    },
    {
      "epoch": 0.052695779458404486,
      "grad_norm": 0.8420175909996033,
      "learning_rate": 2.4868616503611124e-07,
      "loss": 2.0776,
      "step": 216
    },
    {
      "epoch": 0.05293974140034155,
      "grad_norm": 0.737402081489563,
      "learning_rate": 2.486797781591823e-07,
      "loss": 1.9902,
      "step": 217
    },
    {
      "epoch": 0.05318370334227861,
      "grad_norm": 1.4963791370391846,
      "learning_rate": 2.4867338833657884e-07,
      "loss": 2.1924,
      "step": 218
    },
    {
      "epoch": 0.053427665284215664,
      "grad_norm": 0.8707871437072754,
      "learning_rate": 2.4866699556626256e-07,
      "loss": 2.3329,
      "step": 219
    },
    {
      "epoch": 0.05367162722615272,
      "grad_norm": 2.2541556358337402,
      "learning_rate": 2.486605998461933e-07,
      "loss": 2.2628,
      "step": 220
    },
    {
      "epoch": 0.05367162722615272,
      "eval_loss": 2.110635995864868,
      "eval_runtime": 82.4261,
      "eval_samples_per_second": 3.106,
      "eval_steps_per_second": 0.776,
      "step": 220
    },
    {
      "epoch": 0.05391558916808978,
      "grad_norm": 0.7243703603744507,
      "learning_rate": 2.4865420117432884e-07,
      "loss": 2.2738,
      "step": 221
    },
    {
      "epoch": 0.054159551110026835,
      "grad_norm": 2.9058852195739746,
      "learning_rate": 2.4864779954862536e-07,
      "loss": 2.2954,
      "step": 222
    },
    {
      "epoch": 0.05440351305196389,
      "grad_norm": 0.6986073851585388,
      "learning_rate": 2.486413949670369e-07,
      "loss": 2.062,
      "step": 223
    },
    {
      "epoch": 0.05464747499390095,
      "grad_norm": 0.8654193878173828,
      "learning_rate": 2.486349874275158e-07,
      "loss": 1.8318,
      "step": 224
    },
    {
      "epoch": 0.054891436935838006,
      "grad_norm": 1.1562286615371704,
      "learning_rate": 2.486285769280123e-07,
      "loss": 2.3011,
      "step": 225
    },
    {
      "epoch": 0.05513539887777507,
      "grad_norm": 0.9906584024429321,
      "learning_rate": 2.48622163466475e-07,
      "loss": 2.0338,
      "step": 226
    },
    {
      "epoch": 0.05537936081971213,
      "grad_norm": 0.7650761008262634,
      "learning_rate": 2.486157470408504e-07,
      "loss": 2.2399,
      "step": 227
    },
    {
      "epoch": 0.055623322761649184,
      "grad_norm": 0.8273106217384338,
      "learning_rate": 2.4860932764908314e-07,
      "loss": 2.1346,
      "step": 228
    },
    {
      "epoch": 0.05586728470358624,
      "grad_norm": 0.8235612511634827,
      "learning_rate": 2.486029052891161e-07,
      "loss": 2.1117,
      "step": 229
    },
    {
      "epoch": 0.0561112466455233,
      "grad_norm": 0.7551500201225281,
      "learning_rate": 2.4859647995889003e-07,
      "loss": 2.0045,
      "step": 230
    },
    {
      "epoch": 0.0561112466455233,
      "eval_loss": 2.105764150619507,
      "eval_runtime": 82.5574,
      "eval_samples_per_second": 3.101,
      "eval_steps_per_second": 0.775,
      "step": 230
    },
    {
      "epoch": 0.056355208587460355,
      "grad_norm": 0.6984049081802368,
      "learning_rate": 2.4859005165634397e-07,
      "loss": 1.9406,
      "step": 231
    },
    {
      "epoch": 0.05659917052939741,
      "grad_norm": 0.7007377743721008,
      "learning_rate": 2.4858362037941493e-07,
      "loss": 2.1012,
      "step": 232
    },
    {
      "epoch": 0.05684313247133447,
      "grad_norm": 0.7448357343673706,
      "learning_rate": 2.485771861260381e-07,
      "loss": 2.0201,
      "step": 233
    },
    {
      "epoch": 0.05708709441327153,
      "grad_norm": 0.8038643002510071,
      "learning_rate": 2.485707488941467e-07,
      "loss": 2.113,
      "step": 234
    },
    {
      "epoch": 0.05733105635520859,
      "grad_norm": 0.683526873588562,
      "learning_rate": 2.48564308681672e-07,
      "loss": 2.2006,
      "step": 235
    },
    {
      "epoch": 0.05757501829714565,
      "grad_norm": 1.10641610622406,
      "learning_rate": 2.485578654865435e-07,
      "loss": 2.1394,
      "step": 236
    },
    {
      "epoch": 0.057818980239082704,
      "grad_norm": 0.9854748249053955,
      "learning_rate": 2.485514193066886e-07,
      "loss": 2.0461,
      "step": 237
    },
    {
      "epoch": 0.05806294218101976,
      "grad_norm": 0.9628323316574097,
      "learning_rate": 2.485449701400329e-07,
      "loss": 2.0795,
      "step": 238
    },
    {
      "epoch": 0.05830690412295682,
      "grad_norm": 0.7303637862205505,
      "learning_rate": 2.485385179845001e-07,
      "loss": 2.1231,
      "step": 239
    },
    {
      "epoch": 0.058550866064893875,
      "grad_norm": 1.024084448814392,
      "learning_rate": 2.4853206283801187e-07,
      "loss": 2.2952,
      "step": 240
    },
    {
      "epoch": 0.058550866064893875,
      "eval_loss": 2.1005725860595703,
      "eval_runtime": 82.6605,
      "eval_samples_per_second": 3.097,
      "eval_steps_per_second": 0.774,
      "step": 240
    },
    {
      "epoch": 0.05879482800683093,
      "grad_norm": 0.7839118242263794,
      "learning_rate": 2.4852560469848794e-07,
      "loss": 2.1039,
      "step": 241
    },
    {
      "epoch": 0.05903878994876799,
      "grad_norm": 0.8608528971672058,
      "learning_rate": 2.4851914356384624e-07,
      "loss": 2.0228,
      "step": 242
    },
    {
      "epoch": 0.05928275189070505,
      "grad_norm": 0.7604301571846008,
      "learning_rate": 2.485126794320027e-07,
      "loss": 2.085,
      "step": 243
    },
    {
      "epoch": 0.05952671383264211,
      "grad_norm": 0.6948022842407227,
      "learning_rate": 2.4850621230087125e-07,
      "loss": 2.1667,
      "step": 244
    },
    {
      "epoch": 0.05977067577457917,
      "grad_norm": 0.9847801327705383,
      "learning_rate": 2.4849974216836405e-07,
      "loss": 2.132,
      "step": 245
    },
    {
      "epoch": 0.060014637716516224,
      "grad_norm": 0.647404670715332,
      "learning_rate": 2.4849326903239115e-07,
      "loss": 2.0881,
      "step": 246
    },
    {
      "epoch": 0.06025859965845328,
      "grad_norm": 0.8329553604125977,
      "learning_rate": 2.4848679289086074e-07,
      "loss": 2.1774,
      "step": 247
    },
    {
      "epoch": 0.06050256160039034,
      "grad_norm": 0.6948926448822021,
      "learning_rate": 2.4848031374167913e-07,
      "loss": 2.1431,
      "step": 248
    },
    {
      "epoch": 0.060746523542327395,
      "grad_norm": 1.132636547088623,
      "learning_rate": 2.484738315827505e-07,
      "loss": 2.1713,
      "step": 249
    },
    {
      "epoch": 0.06099048548426445,
      "grad_norm": 0.7370600700378418,
      "learning_rate": 2.484673464119773e-07,
      "loss": 2.1122,
      "step": 250
    },
    {
      "epoch": 0.06099048548426445,
      "eval_loss": 2.0957703590393066,
      "eval_runtime": 82.5102,
      "eval_samples_per_second": 3.103,
      "eval_steps_per_second": 0.776,
      "step": 250
    },
    {
      "epoch": 0.061234447426201516,
      "grad_norm": 0.9531499147415161,
      "learning_rate": 2.484608582272598e-07,
      "loss": 2.1948,
      "step": 251
    },
    {
      "epoch": 0.06147840936813857,
      "grad_norm": 49.09546661376953,
      "learning_rate": 2.4845436702649656e-07,
      "loss": 2.3097,
      "step": 252
    },
    {
      "epoch": 0.06172237131007563,
      "grad_norm": 0.9064823389053345,
      "learning_rate": 2.48447872807584e-07,
      "loss": 2.1416,
      "step": 253
    },
    {
      "epoch": 0.06196633325201269,
      "grad_norm": 0.8501002788543701,
      "learning_rate": 2.484413755684167e-07,
      "loss": 2.1266,
      "step": 254
    },
    {
      "epoch": 0.062210295193949744,
      "grad_norm": 0.9969101548194885,
      "learning_rate": 2.484348753068872e-07,
      "loss": 2.1585,
      "step": 255
    },
    {
      "epoch": 0.0624542571358868,
      "grad_norm": 0.7964323163032532,
      "learning_rate": 2.484283720208861e-07,
      "loss": 2.0506,
      "step": 256
    },
    {
      "epoch": 0.06269821907782386,
      "grad_norm": 0.8279253840446472,
      "learning_rate": 2.4842186570830207e-07,
      "loss": 2.0881,
      "step": 257
    },
    {
      "epoch": 0.06294218101976091,
      "grad_norm": 0.751754879951477,
      "learning_rate": 2.484153563670218e-07,
      "loss": 1.93,
      "step": 258
    },
    {
      "epoch": 0.06318614296169797,
      "grad_norm": 0.7673382759094238,
      "learning_rate": 2.4840884399493006e-07,
      "loss": 2.0995,
      "step": 259
    },
    {
      "epoch": 0.06343010490363503,
      "grad_norm": 0.6529645919799805,
      "learning_rate": 2.4840232858990943e-07,
      "loss": 2.0436,
      "step": 260
    },
    {
      "epoch": 0.06343010490363503,
      "eval_loss": 2.0920004844665527,
      "eval_runtime": 82.4308,
      "eval_samples_per_second": 3.106,
      "eval_steps_per_second": 0.776,
      "step": 260
    },
    {
      "epoch": 0.06367406684557209,
      "grad_norm": 1.109803557395935,
      "learning_rate": 2.4839581014984084e-07,
      "loss": 2.1963,
      "step": 261
    },
    {
      "epoch": 0.06391802878750914,
      "grad_norm": 0.8001452088356018,
      "learning_rate": 2.48389288672603e-07,
      "loss": 2.2763,
      "step": 262
    },
    {
      "epoch": 0.0641619907294462,
      "grad_norm": 0.8246157169342041,
      "learning_rate": 2.483827641560728e-07,
      "loss": 2.0909,
      "step": 263
    },
    {
      "epoch": 0.06440595267138327,
      "grad_norm": 1.1484432220458984,
      "learning_rate": 2.48376236598125e-07,
      "loss": 2.0256,
      "step": 264
    },
    {
      "epoch": 0.06464991461332033,
      "grad_norm": 0.7669978737831116,
      "learning_rate": 2.4836970599663255e-07,
      "loss": 2.119,
      "step": 265
    },
    {
      "epoch": 0.06489387655525738,
      "grad_norm": 0.8040069341659546,
      "learning_rate": 2.4836317234946626e-07,
      "loss": 2.2469,
      "step": 266
    },
    {
      "epoch": 0.06513783849719444,
      "grad_norm": 1.2191132307052612,
      "learning_rate": 2.48356635654495e-07,
      "loss": 2.1695,
      "step": 267
    },
    {
      "epoch": 0.0653818004391315,
      "grad_norm": 0.7855357527732849,
      "learning_rate": 2.4835009590958575e-07,
      "loss": 2.165,
      "step": 268
    },
    {
      "epoch": 0.06562576238106856,
      "grad_norm": 0.692392110824585,
      "learning_rate": 2.483435531126034e-07,
      "loss": 2.1611,
      "step": 269
    },
    {
      "epoch": 0.06586972432300561,
      "grad_norm": 0.7394700646400452,
      "learning_rate": 2.483370072614108e-07,
      "loss": 2.0499,
      "step": 270
    },
    {
      "epoch": 0.06586972432300561,
      "eval_loss": 2.0882444381713867,
      "eval_runtime": 82.4623,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 270
    },
    {
      "epoch": 0.06611368626494267,
      "grad_norm": 0.8597064018249512,
      "learning_rate": 2.483304583538689e-07,
      "loss": 2.1721,
      "step": 271
    },
    {
      "epoch": 0.06635764820687973,
      "grad_norm": 0.6948525905609131,
      "learning_rate": 2.4832390638783666e-07,
      "loss": 2.1164,
      "step": 272
    },
    {
      "epoch": 0.06660161014881678,
      "grad_norm": 0.9271346926689148,
      "learning_rate": 2.4831735136117095e-07,
      "loss": 2.222,
      "step": 273
    },
    {
      "epoch": 0.06684557209075384,
      "grad_norm": 0.7124070525169373,
      "learning_rate": 2.4831079327172674e-07,
      "loss": 2.1953,
      "step": 274
    },
    {
      "epoch": 0.0670895340326909,
      "grad_norm": 0.7730213403701782,
      "learning_rate": 2.4830423211735686e-07,
      "loss": 2.1612,
      "step": 275
    },
    {
      "epoch": 0.06733349597462795,
      "grad_norm": 0.7319573760032654,
      "learning_rate": 2.482976678959123e-07,
      "loss": 2.1091,
      "step": 276
    },
    {
      "epoch": 0.06757745791656501,
      "grad_norm": 0.7052650451660156,
      "learning_rate": 2.4829110060524197e-07,
      "loss": 2.0802,
      "step": 277
    },
    {
      "epoch": 0.06782141985850207,
      "grad_norm": 0.6924635171890259,
      "learning_rate": 2.482845302431927e-07,
      "loss": 2.0242,
      "step": 278
    },
    {
      "epoch": 0.06806538180043913,
      "grad_norm": 0.8777127861976624,
      "learning_rate": 2.4827795680760933e-07,
      "loss": 2.1926,
      "step": 279
    },
    {
      "epoch": 0.06830934374237618,
      "grad_norm": 0.6327698230743408,
      "learning_rate": 2.482713802963348e-07,
      "loss": 2.0477,
      "step": 280
    },
    {
      "epoch": 0.06830934374237618,
      "eval_loss": 2.0836799144744873,
      "eval_runtime": 82.5559,
      "eval_samples_per_second": 3.101,
      "eval_steps_per_second": 0.775,
      "step": 280
    },
    {
      "epoch": 0.06855330568431325,
      "grad_norm": 0.8194009065628052,
      "learning_rate": 2.4826480070720985e-07,
      "loss": 2.1662,
      "step": 281
    },
    {
      "epoch": 0.06879726762625031,
      "grad_norm": 2.9831783771514893,
      "learning_rate": 2.482582180380734e-07,
      "loss": 2.0296,
      "step": 282
    },
    {
      "epoch": 0.06904122956818737,
      "grad_norm": 0.7184981107711792,
      "learning_rate": 2.482516322867622e-07,
      "loss": 2.0806,
      "step": 283
    },
    {
      "epoch": 0.06928519151012442,
      "grad_norm": 0.9403382539749146,
      "learning_rate": 2.48245043451111e-07,
      "loss": 2.267,
      "step": 284
    },
    {
      "epoch": 0.06952915345206148,
      "grad_norm": 0.7837559580802917,
      "learning_rate": 2.482384515289525e-07,
      "loss": 2.1853,
      "step": 285
    },
    {
      "epoch": 0.06977311539399854,
      "grad_norm": 0.6912328004837036,
      "learning_rate": 2.482318565181174e-07,
      "loss": 2.1157,
      "step": 286
    },
    {
      "epoch": 0.0700170773359356,
      "grad_norm": 0.7220721244812012,
      "learning_rate": 2.4822525841643453e-07,
      "loss": 2.1604,
      "step": 287
    },
    {
      "epoch": 0.07026103927787265,
      "grad_norm": 0.6910116672515869,
      "learning_rate": 2.482186572217303e-07,
      "loss": 2.3089,
      "step": 288
    },
    {
      "epoch": 0.07050500121980971,
      "grad_norm": 0.6678460836410522,
      "learning_rate": 2.482120529318294e-07,
      "loss": 2.144,
      "step": 289
    },
    {
      "epoch": 0.07074896316174677,
      "grad_norm": 0.8362970948219299,
      "learning_rate": 2.482054455445545e-07,
      "loss": 2.1724,
      "step": 290
    },
    {
      "epoch": 0.07074896316174677,
      "eval_loss": 2.079866409301758,
      "eval_runtime": 82.5785,
      "eval_samples_per_second": 3.1,
      "eval_steps_per_second": 0.775,
      "step": 290
    },
    {
      "epoch": 0.07099292510368382,
      "grad_norm": 0.6900479793548584,
      "learning_rate": 2.481988350577259e-07,
      "loss": 2.0982,
      "step": 291
    },
    {
      "epoch": 0.07123688704562088,
      "grad_norm": 0.737310528755188,
      "learning_rate": 2.481922214691622e-07,
      "loss": 2.0816,
      "step": 292
    },
    {
      "epoch": 0.07148084898755794,
      "grad_norm": 0.6768659949302673,
      "learning_rate": 2.481856047766798e-07,
      "loss": 2.1823,
      "step": 293
    },
    {
      "epoch": 0.071724810929495,
      "grad_norm": 0.6438968181610107,
      "learning_rate": 2.4817898497809304e-07,
      "loss": 2.0001,
      "step": 294
    },
    {
      "epoch": 0.07196877287143205,
      "grad_norm": 0.9217244982719421,
      "learning_rate": 2.4817236207121427e-07,
      "loss": 2.1251,
      "step": 295
    },
    {
      "epoch": 0.07221273481336911,
      "grad_norm": 0.984214186668396,
      "learning_rate": 2.4816573605385374e-07,
      "loss": 2.2588,
      "step": 296
    },
    {
      "epoch": 0.07245669675530617,
      "grad_norm": 0.6501964330673218,
      "learning_rate": 2.481591069238197e-07,
      "loss": 1.9671,
      "step": 297
    },
    {
      "epoch": 0.07270065869724324,
      "grad_norm": 0.7080546617507935,
      "learning_rate": 2.481524746789182e-07,
      "loss": 2.1113,
      "step": 298
    },
    {
      "epoch": 0.0729446206391803,
      "grad_norm": 0.759996235370636,
      "learning_rate": 2.4814583931695343e-07,
      "loss": 2.1671,
      "step": 299
    },
    {
      "epoch": 0.07318858258111735,
      "grad_norm": 0.7538090348243713,
      "learning_rate": 2.4813920083572734e-07,
      "loss": 2.0294,
      "step": 300
    },
    {
      "epoch": 0.07318858258111735,
      "eval_loss": 2.0766212940216064,
      "eval_runtime": 82.5747,
      "eval_samples_per_second": 3.1,
      "eval_steps_per_second": 0.775,
      "step": 300
    },
    {
      "epoch": 0.07343254452305441,
      "grad_norm": 1.0901976823806763,
      "learning_rate": 2.481325592330399e-07,
      "loss": 2.1481,
      "step": 301
    },
    {
      "epoch": 0.07367650646499146,
      "grad_norm": 0.5969391465187073,
      "learning_rate": 2.4812591450668896e-07,
      "loss": 2.0585,
      "step": 302
    },
    {
      "epoch": 0.07392046840692852,
      "grad_norm": 1.0443135499954224,
      "learning_rate": 2.4811926665447034e-07,
      "loss": 2.1669,
      "step": 303
    },
    {
      "epoch": 0.07416443034886558,
      "grad_norm": 0.7203410863876343,
      "learning_rate": 2.481126156741779e-07,
      "loss": 2.1098,
      "step": 304
    },
    {
      "epoch": 0.07440839229080264,
      "grad_norm": 0.8156256079673767,
      "learning_rate": 2.481059615636031e-07,
      "loss": 2.0854,
      "step": 305
    },
    {
      "epoch": 0.07465235423273969,
      "grad_norm": 0.9032734036445618,
      "learning_rate": 2.480993043205356e-07,
      "loss": 2.0709,
      "step": 306
    },
    {
      "epoch": 0.07489631617467675,
      "grad_norm": 0.7365292310714722,
      "learning_rate": 2.4809264394276297e-07,
      "loss": 2.0466,
      "step": 307
    },
    {
      "epoch": 0.0751402781166138,
      "grad_norm": 1.4419785737991333,
      "learning_rate": 2.4808598042807057e-07,
      "loss": 2.1844,
      "step": 308
    },
    {
      "epoch": 0.07538424005855086,
      "grad_norm": 0.8487168550491333,
      "learning_rate": 2.4807931377424167e-07,
      "loss": 2.0844,
      "step": 309
    },
    {
      "epoch": 0.07562820200048792,
      "grad_norm": 0.6521626114845276,
      "learning_rate": 2.4807264397905757e-07,
      "loss": 2.1657,
      "step": 310
    },
    {
      "epoch": 0.07562820200048792,
      "eval_loss": 2.0734453201293945,
      "eval_runtime": 82.6342,
      "eval_samples_per_second": 3.098,
      "eval_steps_per_second": 0.774,
      "step": 310
    },
    {
      "epoch": 0.07587216394242498,
      "grad_norm": 1.0338906049728394,
      "learning_rate": 2.480659710402974e-07,
      "loss": 2.0817,
      "step": 311
    },
    {
      "epoch": 0.07611612588436203,
      "grad_norm": 0.7622800469398499,
      "learning_rate": 2.480592949557383e-07,
      "loss": 2.2028,
      "step": 312
    },
    {
      "epoch": 0.07636008782629909,
      "grad_norm": 0.698598325252533,
      "learning_rate": 2.4805261572315513e-07,
      "loss": 2.1197,
      "step": 313
    },
    {
      "epoch": 0.07660404976823615,
      "grad_norm": 0.9301127195358276,
      "learning_rate": 2.480459333403207e-07,
      "loss": 2.0719,
      "step": 314
    },
    {
      "epoch": 0.07684801171017322,
      "grad_norm": 0.7141032814979553,
      "learning_rate": 2.480392478050059e-07,
      "loss": 1.9979,
      "step": 315
    },
    {
      "epoch": 0.07709197365211028,
      "grad_norm": 0.8011495471000671,
      "learning_rate": 2.4803255911497927e-07,
      "loss": 2.2752,
      "step": 316
    },
    {
      "epoch": 0.07733593559404733,
      "grad_norm": 0.6986822485923767,
      "learning_rate": 2.4802586726800744e-07,
      "loss": 2.1448,
      "step": 317
    },
    {
      "epoch": 0.07757989753598439,
      "grad_norm": 0.7055565118789673,
      "learning_rate": 2.4801917226185476e-07,
      "loss": 2.1929,
      "step": 318
    },
    {
      "epoch": 0.07782385947792145,
      "grad_norm": 0.6355963945388794,
      "learning_rate": 2.480124740942837e-07,
      "loss": 2.0868,
      "step": 319
    },
    {
      "epoch": 0.0780678214198585,
      "grad_norm": 0.6996073126792908,
      "learning_rate": 2.480057727630543e-07,
      "loss": 2.1996,
      "step": 320
    },
    {
      "epoch": 0.0780678214198585,
      "eval_loss": 2.070326805114746,
      "eval_runtime": 82.8377,
      "eval_samples_per_second": 3.09,
      "eval_steps_per_second": 0.773,
      "step": 320
    },
    {
      "epoch": 0.07831178336179556,
      "grad_norm": 0.9807422757148743,
      "learning_rate": 2.479990682659248e-07,
      "loss": 2.1283,
      "step": 321
    },
    {
      "epoch": 0.07855574530373262,
      "grad_norm": 0.660064697265625,
      "learning_rate": 2.4799236060065104e-07,
      "loss": 2.1273,
      "step": 322
    },
    {
      "epoch": 0.07879970724566968,
      "grad_norm": 0.6279289722442627,
      "learning_rate": 2.47985649764987e-07,
      "loss": 2.0789,
      "step": 323
    },
    {
      "epoch": 0.07904366918760673,
      "grad_norm": 0.8380366563796997,
      "learning_rate": 2.4797893575668437e-07,
      "loss": 2.0606,
      "step": 324
    },
    {
      "epoch": 0.07928763112954379,
      "grad_norm": 0.6436744928359985,
      "learning_rate": 2.4797221857349267e-07,
      "loss": 2.1684,
      "step": 325
    },
    {
      "epoch": 0.07953159307148085,
      "grad_norm": 0.9657268524169922,
      "learning_rate": 2.4796549821315954e-07,
      "loss": 2.1487,
      "step": 326
    },
    {
      "epoch": 0.0797755550134179,
      "grad_norm": 0.7531183362007141,
      "learning_rate": 2.479587746734302e-07,
      "loss": 2.2133,
      "step": 327
    },
    {
      "epoch": 0.08001951695535496,
      "grad_norm": 0.6298527121543884,
      "learning_rate": 2.4795204795204794e-07,
      "loss": 2.1123,
      "step": 328
    },
    {
      "epoch": 0.08026347889729202,
      "grad_norm": 1.2394922971725464,
      "learning_rate": 2.479453180467538e-07,
      "loss": 2.1458,
      "step": 329
    },
    {
      "epoch": 0.08050744083922907,
      "grad_norm": 0.7295296788215637,
      "learning_rate": 2.479385849552867e-07,
      "loss": 2.127,
      "step": 330
    },
    {
      "epoch": 0.08050744083922907,
      "eval_loss": 2.0674209594726562,
      "eval_runtime": 82.4919,
      "eval_samples_per_second": 3.103,
      "eval_steps_per_second": 0.776,
      "step": 330
    },
    {
      "epoch": 0.08075140278116613,
      "grad_norm": 0.7010200619697571,
      "learning_rate": 2.479318486753834e-07,
      "loss": 1.9559,
      "step": 331
    },
    {
      "epoch": 0.0809953647231032,
      "grad_norm": 0.9205324053764343,
      "learning_rate": 2.479251092047787e-07,
      "loss": 2.3111,
      "step": 332
    },
    {
      "epoch": 0.08123932666504026,
      "grad_norm": 0.7850201725959778,
      "learning_rate": 2.4791836654120494e-07,
      "loss": 2.0908,
      "step": 333
    },
    {
      "epoch": 0.08148328860697732,
      "grad_norm": 0.8352589011192322,
      "learning_rate": 2.4791162068239256e-07,
      "loss": 2.0713,
      "step": 334
    },
    {
      "epoch": 0.08172725054891437,
      "grad_norm": 1.1444090604782104,
      "learning_rate": 2.4790487162606977e-07,
      "loss": 2.2373,
      "step": 335
    },
    {
      "epoch": 0.08197121249085143,
      "grad_norm": 0.6652919054031372,
      "learning_rate": 2.478981193699626e-07,
      "loss": 2.0393,
      "step": 336
    },
    {
      "epoch": 0.08221517443278849,
      "grad_norm": 1.2522311210632324,
      "learning_rate": 2.478913639117949e-07,
      "loss": 2.1158,
      "step": 337
    },
    {
      "epoch": 0.08245913637472554,
      "grad_norm": 0.6700438261032104,
      "learning_rate": 2.478846052492885e-07,
      "loss": 2.1475,
      "step": 338
    },
    {
      "epoch": 0.0827030983166626,
      "grad_norm": 0.594170868396759,
      "learning_rate": 2.478778433801629e-07,
      "loss": 2.2158,
      "step": 339
    },
    {
      "epoch": 0.08294706025859966,
      "grad_norm": 0.7548016905784607,
      "learning_rate": 2.478710783021355e-07,
      "loss": 2.1094,
      "step": 340
    },
    {
      "epoch": 0.08294706025859966,
      "eval_loss": 2.0647270679473877,
      "eval_runtime": 82.4405,
      "eval_samples_per_second": 3.105,
      "eval_steps_per_second": 0.776,
      "step": 340
    },
    {
      "epoch": 0.08319102220053672,
      "grad_norm": 0.6635969877243042,
      "learning_rate": 2.4786431001292156e-07,
      "loss": 2.0065,
      "step": 341
    },
    {
      "epoch": 0.08343498414247377,
      "grad_norm": 0.5917369723320007,
      "learning_rate": 2.478575385102342e-07,
      "loss": 2.1334,
      "step": 342
    },
    {
      "epoch": 0.08367894608441083,
      "grad_norm": 0.7114012241363525,
      "learning_rate": 2.4785076379178427e-07,
      "loss": 2.2898,
      "step": 343
    },
    {
      "epoch": 0.08392290802634789,
      "grad_norm": 0.6210088729858398,
      "learning_rate": 2.478439858552805e-07,
      "loss": 2.1155,
      "step": 344
    },
    {
      "epoch": 0.08416686996828494,
      "grad_norm": 3.0671420097351074,
      "learning_rate": 2.4783720469842943e-07,
      "loss": 2.2391,
      "step": 345
    },
    {
      "epoch": 0.084410831910222,
      "grad_norm": 0.6841104030609131,
      "learning_rate": 2.4783042031893544e-07,
      "loss": 1.9859,
      "step": 346
    },
    {
      "epoch": 0.08465479385215906,
      "grad_norm": 0.8301260471343994,
      "learning_rate": 2.478236327145007e-07,
      "loss": 2.1063,
      "step": 347
    },
    {
      "epoch": 0.08489875579409611,
      "grad_norm": 0.688525378704071,
      "learning_rate": 2.4781684188282526e-07,
      "loss": 2.0468,
      "step": 348
    },
    {
      "epoch": 0.08514271773603319,
      "grad_norm": 0.7203090190887451,
      "learning_rate": 2.4781004782160693e-07,
      "loss": 2.1424,
      "step": 349
    },
    {
      "epoch": 0.08538667967797024,
      "grad_norm": 0.6987703442573547,
      "learning_rate": 2.478032505285412e-07,
      "loss": 2.0616,
      "step": 350
    },
    {
      "epoch": 0.08538667967797024,
      "eval_loss": 2.0618653297424316,
      "eval_runtime": 82.4978,
      "eval_samples_per_second": 3.103,
      "eval_steps_per_second": 0.776,
      "step": 350
    },
    {
      "epoch": 0.0856306416199073,
      "grad_norm": 0.8080345392227173,
      "learning_rate": 2.4779645000132166e-07,
      "loss": 2.1916,
      "step": 351
    },
    {
      "epoch": 0.08587460356184436,
      "grad_norm": 0.6830054521560669,
      "learning_rate": 2.477896462376395e-07,
      "loss": 2.0308,
      "step": 352
    },
    {
      "epoch": 0.08611856550378141,
      "grad_norm": 0.9455615282058716,
      "learning_rate": 2.4778283923518366e-07,
      "loss": 2.2529,
      "step": 353
    },
    {
      "epoch": 0.08636252744571847,
      "grad_norm": 1.0438265800476074,
      "learning_rate": 2.477760289916411e-07,
      "loss": 2.1359,
      "step": 354
    },
    {
      "epoch": 0.08660648938765553,
      "grad_norm": 0.7835482358932495,
      "learning_rate": 2.477692155046964e-07,
      "loss": 2.2987,
      "step": 355
    },
    {
      "epoch": 0.08685045132959258,
      "grad_norm": 1.003544569015503,
      "learning_rate": 2.47762398772032e-07,
      "loss": 2.2476,
      "step": 356
    },
    {
      "epoch": 0.08709441327152964,
      "grad_norm": 0.7639005780220032,
      "learning_rate": 2.4775557879132803e-07,
      "loss": 2.0089,
      "step": 357
    },
    {
      "epoch": 0.0873383752134667,
      "grad_norm": 0.6183759570121765,
      "learning_rate": 2.4774875556026265e-07,
      "loss": 2.0263,
      "step": 358
    },
    {
      "epoch": 0.08758233715540376,
      "grad_norm": 0.9802903532981873,
      "learning_rate": 2.477419290765115e-07,
      "loss": 2.0125,
      "step": 359
    },
    {
      "epoch": 0.08782629909734081,
      "grad_norm": 0.7445787787437439,
      "learning_rate": 2.4773509933774833e-07,
      "loss": 1.9287,
      "step": 360
    },
    {
      "epoch": 0.08782629909734081,
      "eval_loss": 2.060185194015503,
      "eval_runtime": 82.5332,
      "eval_samples_per_second": 3.102,
      "eval_steps_per_second": 0.775,
      "step": 360
    },
    {
      "epoch": 0.08807026103927787,
      "grad_norm": 0.6245632767677307,
      "learning_rate": 2.4772826634164435e-07,
      "loss": 1.9036,
      "step": 361
    },
    {
      "epoch": 0.08831422298121493,
      "grad_norm": 0.822571873664856,
      "learning_rate": 2.4772143008586876e-07,
      "loss": 2.1521,
      "step": 362
    },
    {
      "epoch": 0.08855818492315198,
      "grad_norm": 0.7154176235198975,
      "learning_rate": 2.4771459056808844e-07,
      "loss": 2.1507,
      "step": 363
    },
    {
      "epoch": 0.08880214686508904,
      "grad_norm": 0.7716903686523438,
      "learning_rate": 2.477077477859681e-07,
      "loss": 2.0469,
      "step": 364
    },
    {
      "epoch": 0.0890461088070261,
      "grad_norm": 0.8829085230827332,
      "learning_rate": 2.4770090173717014e-07,
      "loss": 2.0799,
      "step": 365
    },
    {
      "epoch": 0.08929007074896317,
      "grad_norm": 0.6315395832061768,
      "learning_rate": 2.4769405241935484e-07,
      "loss": 2.0283,
      "step": 366
    },
    {
      "epoch": 0.08953403269090023,
      "grad_norm": 0.6415708661079407,
      "learning_rate": 2.476871998301802e-07,
      "loss": 2.0164,
      "step": 367
    },
    {
      "epoch": 0.08977799463283728,
      "grad_norm": 0.6306458115577698,
      "learning_rate": 2.476803439673019e-07,
      "loss": 2.1085,
      "step": 368
    },
    {
      "epoch": 0.09002195657477434,
      "grad_norm": 0.8052152991294861,
      "learning_rate": 2.476734848283735e-07,
      "loss": 2.1536,
      "step": 369
    },
    {
      "epoch": 0.0902659185167114,
      "grad_norm": 0.7443030476570129,
      "learning_rate": 2.476666224110462e-07,
      "loss": 2.2221,
      "step": 370
    },
    {
      "epoch": 0.0902659185167114,
      "eval_loss": 2.057771682739258,
      "eval_runtime": 82.4526,
      "eval_samples_per_second": 3.105,
      "eval_steps_per_second": 0.776,
      "step": 370
    },
    {
      "epoch": 0.09050988045864845,
      "grad_norm": 0.717291533946991,
      "learning_rate": 2.476597567129691e-07,
      "loss": 1.9786,
      "step": 371
    },
    {
      "epoch": 0.09075384240058551,
      "grad_norm": 0.7804278135299683,
      "learning_rate": 2.4765288773178894e-07,
      "loss": 2.1336,
      "step": 372
    },
    {
      "epoch": 0.09099780434252257,
      "grad_norm": 0.5851336121559143,
      "learning_rate": 2.476460154651503e-07,
      "loss": 2.0796,
      "step": 373
    },
    {
      "epoch": 0.09124176628445962,
      "grad_norm": 0.7063292860984802,
      "learning_rate": 2.4763913991069527e-07,
      "loss": 2.1067,
      "step": 374
    },
    {
      "epoch": 0.09148572822639668,
      "grad_norm": 0.6482515335083008,
      "learning_rate": 2.4763226106606407e-07,
      "loss": 2.0642,
      "step": 375
    },
    {
      "epoch": 0.09172969016833374,
      "grad_norm": 1.2077556848526,
      "learning_rate": 2.476253789288943e-07,
      "loss": 1.9154,
      "step": 376
    },
    {
      "epoch": 0.0919736521102708,
      "grad_norm": 0.8584334254264832,
      "learning_rate": 2.4761849349682154e-07,
      "loss": 1.9837,
      "step": 377
    },
    {
      "epoch": 0.09221761405220785,
      "grad_norm": 0.5971949696540833,
      "learning_rate": 2.4761160476747895e-07,
      "loss": 1.9219,
      "step": 378
    },
    {
      "epoch": 0.09246157599414491,
      "grad_norm": 1.1772984266281128,
      "learning_rate": 2.4760471273849755e-07,
      "loss": 2.1976,
      "step": 379
    },
    {
      "epoch": 0.09270553793608197,
      "grad_norm": 1.0061829090118408,
      "learning_rate": 2.47597817407506e-07,
      "loss": 2.1878,
      "step": 380
    },
    {
      "epoch": 0.09270553793608197,
      "eval_loss": 2.0550973415374756,
      "eval_runtime": 82.389,
      "eval_samples_per_second": 3.107,
      "eval_steps_per_second": 0.777,
      "step": 380
    },
    {
      "epoch": 0.09294949987801902,
      "grad_norm": 89.18840789794922,
      "learning_rate": 2.475909187721307e-07,
      "loss": 2.1648,
      "step": 381
    },
    {
      "epoch": 0.09319346181995608,
      "grad_norm": 1.159109115600586,
      "learning_rate": 2.4758401682999573e-07,
      "loss": 2.1707,
      "step": 382
    },
    {
      "epoch": 0.09343742376189315,
      "grad_norm": 0.7915341258049011,
      "learning_rate": 2.475771115787231e-07,
      "loss": 2.11,
      "step": 383
    },
    {
      "epoch": 0.09368138570383021,
      "grad_norm": 0.691881537437439,
      "learning_rate": 2.475702030159322e-07,
      "loss": 1.9882,
      "step": 384
    },
    {
      "epoch": 0.09392534764576727,
      "grad_norm": 1.018052577972412,
      "learning_rate": 2.475632911392405e-07,
      "loss": 2.0549,
      "step": 385
    },
    {
      "epoch": 0.09416930958770432,
      "grad_norm": 0.9836434721946716,
      "learning_rate": 2.475563759462629e-07,
      "loss": 2.177,
      "step": 386
    },
    {
      "epoch": 0.09441327152964138,
      "grad_norm": 0.7274807691574097,
      "learning_rate": 2.475494574346122e-07,
      "loss": 2.1897,
      "step": 387
    },
    {
      "epoch": 0.09465723347157844,
      "grad_norm": 0.853111982345581,
      "learning_rate": 2.475425356018988e-07,
      "loss": 2.0191,
      "step": 388
    },
    {
      "epoch": 0.0949011954135155,
      "grad_norm": 0.7096747159957886,
      "learning_rate": 2.475356104457307e-07,
      "loss": 2.0043,
      "step": 389
    },
    {
      "epoch": 0.09514515735545255,
      "grad_norm": 0.59073406457901,
      "learning_rate": 2.4752868196371393e-07,
      "loss": 2.1815,
      "step": 390
    },
    {
      "epoch": 0.09514515735545255,
      "eval_loss": 2.0534627437591553,
      "eval_runtime": 82.449,
      "eval_samples_per_second": 3.105,
      "eval_steps_per_second": 0.776,
      "step": 390
    },
    {
      "epoch": 0.09538911929738961,
      "grad_norm": 1.0871241092681885,
      "learning_rate": 2.47521750153452e-07,
      "loss": 2.3593,
      "step": 391
    },
    {
      "epoch": 0.09563308123932666,
      "grad_norm": 0.7196955680847168,
      "learning_rate": 2.4751481501254606e-07,
      "loss": 2.1606,
      "step": 392
    },
    {
      "epoch": 0.09587704318126372,
      "grad_norm": 0.6455821394920349,
      "learning_rate": 2.4750787653859505e-07,
      "loss": 2.1609,
      "step": 393
    },
    {
      "epoch": 0.09612100512320078,
      "grad_norm": 0.8399761915206909,
      "learning_rate": 2.475009347291956e-07,
      "loss": 2.2308,
      "step": 394
    },
    {
      "epoch": 0.09636496706513784,
      "grad_norm": 0.7365739941596985,
      "learning_rate": 2.47493989581942e-07,
      "loss": 2.1575,
      "step": 395
    },
    {
      "epoch": 0.09660892900707489,
      "grad_norm": 0.7569345235824585,
      "learning_rate": 2.4748704109442635e-07,
      "loss": 2.1495,
      "step": 396
    },
    {
      "epoch": 0.09685289094901195,
      "grad_norm": 0.6441726088523865,
      "learning_rate": 2.4748008926423817e-07,
      "loss": 2.0264,
      "step": 397
    },
    {
      "epoch": 0.097096852890949,
      "grad_norm": 0.6600158214569092,
      "learning_rate": 2.474731340889649e-07,
      "loss": 2.1404,
      "step": 398
    },
    {
      "epoch": 0.09734081483288606,
      "grad_norm": 0.6377738118171692,
      "learning_rate": 2.4746617556619163e-07,
      "loss": 2.0164,
      "step": 399
    },
    {
      "epoch": 0.09758477677482313,
      "grad_norm": 0.7105040550231934,
      "learning_rate": 2.4745921369350094e-07,
      "loss": 2.1193,
      "step": 400
    },
    {
      "epoch": 0.09758477677482313,
      "eval_loss": 2.051236391067505,
      "eval_runtime": 82.3788,
      "eval_samples_per_second": 3.108,
      "eval_steps_per_second": 0.777,
      "step": 400
    },
    {
      "epoch": 0.09782873871676019,
      "grad_norm": 0.9596676230430603,
      "learning_rate": 2.474522484684733e-07,
      "loss": 2.1332,
      "step": 401
    },
    {
      "epoch": 0.09807270065869725,
      "grad_norm": 0.7559407353401184,
      "learning_rate": 2.4744527988868673e-07,
      "loss": 2.0184,
      "step": 402
    },
    {
      "epoch": 0.0983166626006343,
      "grad_norm": 0.9651991128921509,
      "learning_rate": 2.4743830795171695e-07,
      "loss": 2.0775,
      "step": 403
    },
    {
      "epoch": 0.09856062454257136,
      "grad_norm": 0.6360098719596863,
      "learning_rate": 2.474313326551373e-07,
      "loss": 1.9806,
      "step": 404
    },
    {
      "epoch": 0.09880458648450842,
      "grad_norm": 0.8414661288261414,
      "learning_rate": 2.474243539965189e-07,
      "loss": 2.0598,
      "step": 405
    },
    {
      "epoch": 0.09904854842644548,
      "grad_norm": 0.593665361404419,
      "learning_rate": 2.4741737197343045e-07,
      "loss": 2.1291,
      "step": 406
    },
    {
      "epoch": 0.09929251036838253,
      "grad_norm": 0.6151170134544373,
      "learning_rate": 2.4741038658343824e-07,
      "loss": 2.0288,
      "step": 407
    },
    {
      "epoch": 0.09953647231031959,
      "grad_norm": 0.8486893773078918,
      "learning_rate": 2.474033978241063e-07,
      "loss": 2.1914,
      "step": 408
    },
    {
      "epoch": 0.09978043425225665,
      "grad_norm": 0.5917589664459229,
      "learning_rate": 2.4739640569299634e-07,
      "loss": 1.997,
      "step": 409
    },
    {
      "epoch": 0.1000243961941937,
      "grad_norm": 0.6412562727928162,
      "learning_rate": 2.4738941018766753e-07,
      "loss": 2.1638,
      "step": 410
    },
    {
      "epoch": 0.1000243961941937,
      "eval_loss": 2.0491929054260254,
      "eval_runtime": 82.4035,
      "eval_samples_per_second": 3.107,
      "eval_steps_per_second": 0.777,
      "step": 410
    },
    {
      "epoch": 0.10026835813613076,
      "grad_norm": 0.8086695075035095,
      "learning_rate": 2.47382411305677e-07,
      "loss": 2.0884,
      "step": 411
    },
    {
      "epoch": 0.10051232007806782,
      "grad_norm": 0.7479122281074524,
      "learning_rate": 2.4737540904457914e-07,
      "loss": 2.1612,
      "step": 412
    },
    {
      "epoch": 0.10075628202000488,
      "grad_norm": 0.6791818141937256,
      "learning_rate": 2.4736840340192635e-07,
      "loss": 2.0849,
      "step": 413
    },
    {
      "epoch": 0.10100024396194193,
      "grad_norm": 0.7554202675819397,
      "learning_rate": 2.4736139437526835e-07,
      "loss": 2.0364,
      "step": 414
    },
    {
      "epoch": 0.10124420590387899,
      "grad_norm": 1.182301640510559,
      "learning_rate": 2.4735438196215273e-07,
      "loss": 2.0642,
      "step": 415
    },
    {
      "epoch": 0.10148816784581605,
      "grad_norm": 0.7850328087806702,
      "learning_rate": 2.4734736616012457e-07,
      "loss": 2.0743,
      "step": 416
    },
    {
      "epoch": 0.10173212978775312,
      "grad_norm": 0.8438737988471985,
      "learning_rate": 2.4734034696672667e-07,
      "loss": 2.1803,
      "step": 417
    },
    {
      "epoch": 0.10197609172969017,
      "grad_norm": 5.938900947570801,
      "learning_rate": 2.473333243794993e-07,
      "loss": 2.1007,
      "step": 418
    },
    {
      "epoch": 0.10222005367162723,
      "grad_norm": 0.7063493728637695,
      "learning_rate": 2.4732629839598054e-07,
      "loss": 2.0102,
      "step": 419
    },
    {
      "epoch": 0.10246401561356429,
      "grad_norm": 0.7309712171554565,
      "learning_rate": 2.4731926901370596e-07,
      "loss": 2.0201,
      "step": 420
    },
    {
      "epoch": 0.10246401561356429,
      "eval_loss": 2.047513484954834,
      "eval_runtime": 82.2828,
      "eval_samples_per_second": 3.111,
      "eval_steps_per_second": 0.778,
      "step": 420
    },
    {
      "epoch": 0.10270797755550135,
      "grad_norm": 0.6622780561447144,
      "learning_rate": 2.473122362302088e-07,
      "loss": 2.1224,
      "step": 421
    },
    {
      "epoch": 0.1029519394974384,
      "grad_norm": 0.5755351185798645,
      "learning_rate": 2.4730520004301997e-07,
      "loss": 2.0315,
      "step": 422
    },
    {
      "epoch": 0.10319590143937546,
      "grad_norm": 0.6571397185325623,
      "learning_rate": 2.472981604496678e-07,
      "loss": 2.1613,
      "step": 423
    },
    {
      "epoch": 0.10343986338131252,
      "grad_norm": 0.6562328338623047,
      "learning_rate": 2.472911174476784e-07,
      "loss": 2.1833,
      "step": 424
    },
    {
      "epoch": 0.10368382532324957,
      "grad_norm": 0.6494048237800598,
      "learning_rate": 2.4728407103457554e-07,
      "loss": 2.0679,
      "step": 425
    },
    {
      "epoch": 0.10392778726518663,
      "grad_norm": 1.0681666135787964,
      "learning_rate": 2.472770212078803e-07,
      "loss": 2.1321,
      "step": 426
    },
    {
      "epoch": 0.10417174920712369,
      "grad_norm": 0.7042210698127747,
      "learning_rate": 2.4726996796511157e-07,
      "loss": 2.1927,
      "step": 427
    },
    {
      "epoch": 0.10441571114906074,
      "grad_norm": 0.6480265259742737,
      "learning_rate": 2.4726291130378586e-07,
      "loss": 2.3393,
      "step": 428
    },
    {
      "epoch": 0.1046596730909978,
      "grad_norm": 0.6662938594818115,
      "learning_rate": 2.472558512214172e-07,
      "loss": 2.0135,
      "step": 429
    },
    {
      "epoch": 0.10490363503293486,
      "grad_norm": 0.6291259527206421,
      "learning_rate": 2.4724878771551725e-07,
      "loss": 2.1968,
      "step": 430
    },
    {
      "epoch": 0.10490363503293486,
      "eval_loss": 2.0453972816467285,
      "eval_runtime": 82.4758,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 430
    },
    {
      "epoch": 0.10514759697487192,
      "grad_norm": 3.447300910949707,
      "learning_rate": 2.4724172078359513e-07,
      "loss": 2.0465,
      "step": 431
    },
    {
      "epoch": 0.10539155891680897,
      "grad_norm": 1.3743339776992798,
      "learning_rate": 2.4723465042315776e-07,
      "loss": 2.0544,
      "step": 432
    },
    {
      "epoch": 0.10563552085874603,
      "grad_norm": 0.8059394359588623,
      "learning_rate": 2.4722757663170946e-07,
      "loss": 2.1522,
      "step": 433
    },
    {
      "epoch": 0.1058794828006831,
      "grad_norm": 0.6414276361465454,
      "learning_rate": 2.4722049940675223e-07,
      "loss": 2.1575,
      "step": 434
    },
    {
      "epoch": 0.10612344474262016,
      "grad_norm": 1.0368949174880981,
      "learning_rate": 2.472134187457856e-07,
      "loss": 1.9911,
      "step": 435
    },
    {
      "epoch": 0.10636740668455721,
      "grad_norm": 0.6836427450180054,
      "learning_rate": 2.4720633464630656e-07,
      "loss": 2.074,
      "step": 436
    },
    {
      "epoch": 0.10661136862649427,
      "grad_norm": 0.6823791861534119,
      "learning_rate": 2.4719924710581e-07,
      "loss": 2.0613,
      "step": 437
    },
    {
      "epoch": 0.10685533056843133,
      "grad_norm": 0.6797323822975159,
      "learning_rate": 2.4719215612178795e-07,
      "loss": 2.0521,
      "step": 438
    },
    {
      "epoch": 0.10709929251036839,
      "grad_norm": 0.6761137247085571,
      "learning_rate": 2.471850616917303e-07,
      "loss": 2.0434,
      "step": 439
    },
    {
      "epoch": 0.10734325445230544,
      "grad_norm": 0.6031590104103088,
      "learning_rate": 2.4717796381312446e-07,
      "loss": 2.1448,
      "step": 440
    },
    {
      "epoch": 0.10734325445230544,
      "eval_loss": 2.0439488887786865,
      "eval_runtime": 82.3918,
      "eval_samples_per_second": 3.107,
      "eval_steps_per_second": 0.777,
      "step": 440
    },
    {
      "epoch": 0.1075872163942425,
      "grad_norm": 0.7096999287605286,
      "learning_rate": 2.471708624834553e-07,
      "loss": 2.236,
      "step": 441
    },
    {
      "epoch": 0.10783117833617956,
      "grad_norm": 0.7802643775939941,
      "learning_rate": 2.471637577002053e-07,
      "loss": 2.0246,
      "step": 442
    },
    {
      "epoch": 0.10807514027811661,
      "grad_norm": 0.8068674802780151,
      "learning_rate": 2.471566494608545e-07,
      "loss": 2.1481,
      "step": 443
    },
    {
      "epoch": 0.10831910222005367,
      "grad_norm": 0.792898416519165,
      "learning_rate": 2.4714953776288044e-07,
      "loss": 1.9699,
      "step": 444
    },
    {
      "epoch": 0.10856306416199073,
      "grad_norm": 0.6124379634857178,
      "learning_rate": 2.471424226037583e-07,
      "loss": 2.1199,
      "step": 445
    },
    {
      "epoch": 0.10880702610392778,
      "grad_norm": 1.238782525062561,
      "learning_rate": 2.471353039809606e-07,
      "loss": 2.1193,
      "step": 446
    },
    {
      "epoch": 0.10905098804586484,
      "grad_norm": 0.5847099423408508,
      "learning_rate": 2.471281818919577e-07,
      "loss": 2.2197,
      "step": 447
    },
    {
      "epoch": 0.1092949499878019,
      "grad_norm": 0.7409020066261292,
      "learning_rate": 2.4712105633421726e-07,
      "loss": 2.1878,
      "step": 448
    },
    {
      "epoch": 0.10953891192973896,
      "grad_norm": 0.8623873591423035,
      "learning_rate": 2.471139273052045e-07,
      "loss": 2.102,
      "step": 449
    },
    {
      "epoch": 0.10978287387167601,
      "grad_norm": 0.9536486864089966,
      "learning_rate": 2.471067948023822e-07,
      "loss": 2.0028,
      "step": 450
    },
    {
      "epoch": 0.10978287387167601,
      "eval_loss": 2.042297601699829,
      "eval_runtime": 82.3539,
      "eval_samples_per_second": 3.109,
      "eval_steps_per_second": 0.777,
      "step": 450
    },
    {
      "epoch": 0.11002683581361308,
      "grad_norm": 2.532158136367798,
      "learning_rate": 2.4709965882321085e-07,
      "loss": 2.1556,
      "step": 451
    },
    {
      "epoch": 0.11027079775555014,
      "grad_norm": 0.7216641306877136,
      "learning_rate": 2.470925193651481e-07,
      "loss": 2.2463,
      "step": 452
    },
    {
      "epoch": 0.1105147596974872,
      "grad_norm": 0.7760343551635742,
      "learning_rate": 2.470853764256495e-07,
      "loss": 2.102,
      "step": 453
    },
    {
      "epoch": 0.11075872163942425,
      "grad_norm": 0.8911886811256409,
      "learning_rate": 2.4707823000216777e-07,
      "loss": 2.3057,
      "step": 454
    },
    {
      "epoch": 0.11100268358136131,
      "grad_norm": 0.7775437235832214,
      "learning_rate": 2.470710800921534e-07,
      "loss": 2.0745,
      "step": 455
    },
    {
      "epoch": 0.11124664552329837,
      "grad_norm": 0.6954789757728577,
      "learning_rate": 2.470639266930543e-07,
      "loss": 1.9749,
      "step": 456
    },
    {
      "epoch": 0.11149060746523543,
      "grad_norm": 0.5103456974029541,
      "learning_rate": 2.4705676980231577e-07,
      "loss": 2.053,
      "step": 457
    },
    {
      "epoch": 0.11173456940717248,
      "grad_norm": 0.9708994626998901,
      "learning_rate": 2.4704960941738093e-07,
      "loss": 2.0905,
      "step": 458
    },
    {
      "epoch": 0.11197853134910954,
      "grad_norm": 0.7622889876365662,
      "learning_rate": 2.4704244553569005e-07,
      "loss": 2.2957,
      "step": 459
    },
    {
      "epoch": 0.1122224932910466,
      "grad_norm": 0.7484825849533081,
      "learning_rate": 2.470352781546811e-07,
      "loss": 2.0953,
      "step": 460
    },
    {
      "epoch": 0.1122224932910466,
      "eval_loss": 2.0406336784362793,
      "eval_runtime": 82.4811,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 460
    },
    {
      "epoch": 0.11246645523298365,
      "grad_norm": 0.9012355208396912,
      "learning_rate": 2.4702810727178955e-07,
      "loss": 2.092,
      "step": 461
    },
    {
      "epoch": 0.11271041717492071,
      "grad_norm": 0.9759025573730469,
      "learning_rate": 2.470209328844483e-07,
      "loss": 2.2126,
      "step": 462
    },
    {
      "epoch": 0.11295437911685777,
      "grad_norm": 0.6701992154121399,
      "learning_rate": 2.470137549900877e-07,
      "loss": 2.0805,
      "step": 463
    },
    {
      "epoch": 0.11319834105879482,
      "grad_norm": 0.9370368123054504,
      "learning_rate": 2.4700657358613573e-07,
      "loss": 2.1702,
      "step": 464
    },
    {
      "epoch": 0.11344230300073188,
      "grad_norm": 0.7537663578987122,
      "learning_rate": 2.4699938867001765e-07,
      "loss": 2.1348,
      "step": 465
    },
    {
      "epoch": 0.11368626494266894,
      "grad_norm": 28.197704315185547,
      "learning_rate": 2.469922002391564e-07,
      "loss": 2.1985,
      "step": 466
    },
    {
      "epoch": 0.113930226884606,
      "grad_norm": 0.7190595269203186,
      "learning_rate": 2.4698500829097235e-07,
      "loss": 2.065,
      "step": 467
    },
    {
      "epoch": 0.11417418882654307,
      "grad_norm": 0.6859692931175232,
      "learning_rate": 2.469778128228832e-07,
      "loss": 2.2628,
      "step": 468
    },
    {
      "epoch": 0.11441815076848012,
      "grad_norm": 1.0769257545471191,
      "learning_rate": 2.4697061383230436e-07,
      "loss": 2.099,
      "step": 469
    },
    {
      "epoch": 0.11466211271041718,
      "grad_norm": 0.7370059490203857,
      "learning_rate": 2.469634113166485e-07,
      "loss": 2.0264,
      "step": 470
    },
    {
      "epoch": 0.11466211271041718,
      "eval_loss": 2.03948712348938,
      "eval_runtime": 82.5195,
      "eval_samples_per_second": 3.102,
      "eval_steps_per_second": 0.776,
      "step": 470
    },
    {
      "epoch": 0.11490607465235424,
      "grad_norm": 1.2912418842315674,
      "learning_rate": 2.4695620527332587e-07,
      "loss": 1.9958,
      "step": 471
    },
    {
      "epoch": 0.1151500365942913,
      "grad_norm": 0.7174116373062134,
      "learning_rate": 2.4694899569974417e-07,
      "loss": 2.1067,
      "step": 472
    },
    {
      "epoch": 0.11539399853622835,
      "grad_norm": 0.8682342171669006,
      "learning_rate": 2.4694178259330843e-07,
      "loss": 2.0431,
      "step": 473
    },
    {
      "epoch": 0.11563796047816541,
      "grad_norm": 0.7805954217910767,
      "learning_rate": 2.4693456595142144e-07,
      "loss": 1.936,
      "step": 474
    },
    {
      "epoch": 0.11588192242010247,
      "grad_norm": 0.6474030017852783,
      "learning_rate": 2.46927345771483e-07,
      "loss": 1.9181,
      "step": 475
    },
    {
      "epoch": 0.11612588436203952,
      "grad_norm": 1.033121943473816,
      "learning_rate": 2.4692012205089086e-07,
      "loss": 2.1393,
      "step": 476
    },
    {
      "epoch": 0.11636984630397658,
      "grad_norm": 0.6728079915046692,
      "learning_rate": 2.469128947870398e-07,
      "loss": 2.0636,
      "step": 477
    },
    {
      "epoch": 0.11661380824591364,
      "grad_norm": 0.7054829001426697,
      "learning_rate": 2.4690566397732225e-07,
      "loss": 2.0255,
      "step": 478
    },
    {
      "epoch": 0.11685777018785069,
      "grad_norm": 0.6209118962287903,
      "learning_rate": 2.4689842961912813e-07,
      "loss": 2.2017,
      "step": 479
    },
    {
      "epoch": 0.11710173212978775,
      "grad_norm": 0.8064582943916321,
      "learning_rate": 2.4689119170984457e-07,
      "loss": 2.0825,
      "step": 480
    },
    {
      "epoch": 0.11710173212978775,
      "eval_loss": 2.0377635955810547,
      "eval_runtime": 82.47,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 480
    },
    {
      "epoch": 0.11734569407172481,
      "grad_norm": 0.6298145651817322,
      "learning_rate": 2.4688395024685635e-07,
      "loss": 1.9996,
      "step": 481
    },
    {
      "epoch": 0.11758965601366186,
      "grad_norm": 0.6311830282211304,
      "learning_rate": 2.4687670522754556e-07,
      "loss": 2.025,
      "step": 482
    },
    {
      "epoch": 0.11783361795559892,
      "grad_norm": 0.7297334671020508,
      "learning_rate": 2.468694566492918e-07,
      "loss": 2.2315,
      "step": 483
    },
    {
      "epoch": 0.11807757989753598,
      "grad_norm": 0.8511336445808411,
      "learning_rate": 2.468622045094721e-07,
      "loss": 2.0218,
      "step": 484
    },
    {
      "epoch": 0.11832154183947305,
      "grad_norm": 1.0287209749221802,
      "learning_rate": 2.4685494880546075e-07,
      "loss": 2.0803,
      "step": 485
    },
    {
      "epoch": 0.1185655037814101,
      "grad_norm": 1.0278816223144531,
      "learning_rate": 2.468476895346296e-07,
      "loss": 2.0773,
      "step": 486
    },
    {
      "epoch": 0.11880946572334716,
      "grad_norm": 0.6326528787612915,
      "learning_rate": 2.468404266943481e-07,
      "loss": 1.9748,
      "step": 487
    },
    {
      "epoch": 0.11905342766528422,
      "grad_norm": 0.7467640042304993,
      "learning_rate": 2.4683316028198264e-07,
      "loss": 2.1406,
      "step": 488
    },
    {
      "epoch": 0.11929738960722128,
      "grad_norm": 2.9902219772338867,
      "learning_rate": 2.4682589029489734e-07,
      "loss": 2.2492,
      "step": 489
    },
    {
      "epoch": 0.11954135154915833,
      "grad_norm": 0.7884283661842346,
      "learning_rate": 2.468186167304538e-07,
      "loss": 2.125,
      "step": 490
    },
    {
      "epoch": 0.11954135154915833,
      "eval_loss": 2.0366787910461426,
      "eval_runtime": 82.4749,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 490
    },
    {
      "epoch": 0.11978531349109539,
      "grad_norm": 0.8623523116111755,
      "learning_rate": 2.4681133958601076e-07,
      "loss": 2.0296,
      "step": 491
    },
    {
      "epoch": 0.12002927543303245,
      "grad_norm": 1.98282790184021,
      "learning_rate": 2.4680405885892456e-07,
      "loss": 1.9251,
      "step": 492
    },
    {
      "epoch": 0.1202732373749695,
      "grad_norm": 0.6307066082954407,
      "learning_rate": 2.4679677454654887e-07,
      "loss": 2.0975,
      "step": 493
    },
    {
      "epoch": 0.12051719931690656,
      "grad_norm": 0.6835567951202393,
      "learning_rate": 2.4678948664623467e-07,
      "loss": 2.0954,
      "step": 494
    },
    {
      "epoch": 0.12076116125884362,
      "grad_norm": 0.6431268453598022,
      "learning_rate": 2.467821951553305e-07,
      "loss": 1.9843,
      "step": 495
    },
    {
      "epoch": 0.12100512320078068,
      "grad_norm": 0.6810166835784912,
      "learning_rate": 2.467749000711822e-07,
      "loss": 1.9317,
      "step": 496
    },
    {
      "epoch": 0.12124908514271773,
      "grad_norm": 0.6672168970108032,
      "learning_rate": 2.467676013911329e-07,
      "loss": 1.8837,
      "step": 497
    },
    {
      "epoch": 0.12149304708465479,
      "grad_norm": 0.6653776168823242,
      "learning_rate": 2.467602991125233e-07,
      "loss": 2.1792,
      "step": 498
    },
    {
      "epoch": 0.12173700902659185,
      "grad_norm": 0.7671879529953003,
      "learning_rate": 2.467529932326913e-07,
      "loss": 2.0863,
      "step": 499
    },
    {
      "epoch": 0.1219809709685289,
      "grad_norm": 0.7360325455665588,
      "learning_rate": 2.467456837489723e-07,
      "loss": 2.2625,
      "step": 500
    },
    {
      "epoch": 0.1219809709685289,
      "eval_loss": 2.0352420806884766,
      "eval_runtime": 82.4773,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 500
    },
    {
      "epoch": 0.12222493291046596,
      "grad_norm": 1.1433802843093872,
      "learning_rate": 2.46738370658699e-07,
      "loss": 2.0469,
      "step": 501
    },
    {
      "epoch": 0.12246889485240303,
      "grad_norm": 0.774438738822937,
      "learning_rate": 2.467310539592016e-07,
      "loss": 2.159,
      "step": 502
    },
    {
      "epoch": 0.12271285679434009,
      "grad_norm": 0.5934178233146667,
      "learning_rate": 2.467237336478074e-07,
      "loss": 2.1611,
      "step": 503
    },
    {
      "epoch": 0.12295681873627715,
      "grad_norm": 0.76198810338974,
      "learning_rate": 2.4671640972184124e-07,
      "loss": 2.065,
      "step": 504
    },
    {
      "epoch": 0.1232007806782142,
      "grad_norm": 2.332010507583618,
      "learning_rate": 2.4670908217862535e-07,
      "loss": 2.2185,
      "step": 505
    },
    {
      "epoch": 0.12344474262015126,
      "grad_norm": 0.590925395488739,
      "learning_rate": 2.4670175101547916e-07,
      "loss": 1.9858,
      "step": 506
    },
    {
      "epoch": 0.12368870456208832,
      "grad_norm": 0.6301215887069702,
      "learning_rate": 2.466944162297197e-07,
      "loss": 2.0693,
      "step": 507
    },
    {
      "epoch": 0.12393266650402537,
      "grad_norm": 0.6086316704750061,
      "learning_rate": 2.466870778186611e-07,
      "loss": 2.1286,
      "step": 508
    },
    {
      "epoch": 0.12417662844596243,
      "grad_norm": 0.5990573167800903,
      "learning_rate": 2.466797357796149e-07,
      "loss": 1.9842,
      "step": 509
    },
    {
      "epoch": 0.12442059038789949,
      "grad_norm": 0.7765055894851685,
      "learning_rate": 2.466723901098901e-07,
      "loss": 2.0005,
      "step": 510
    },
    {
      "epoch": 0.12442059038789949,
      "eval_loss": 2.033975124359131,
      "eval_runtime": 82.4194,
      "eval_samples_per_second": 3.106,
      "eval_steps_per_second": 0.777,
      "step": 510
    },
    {
      "epoch": 0.12466455232983654,
      "grad_norm": 0.7282994389533997,
      "learning_rate": 2.466650408067929e-07,
      "loss": 2.0578,
      "step": 511
    },
    {
      "epoch": 0.1249085142717736,
      "grad_norm": 0.6947245597839355,
      "learning_rate": 2.4665768786762685e-07,
      "loss": 2.1235,
      "step": 512
    },
    {
      "epoch": 0.12515247621371067,
      "grad_norm": 1.0526394844055176,
      "learning_rate": 2.4665033128969293e-07,
      "loss": 2.0394,
      "step": 513
    },
    {
      "epoch": 0.12539643815564772,
      "grad_norm": 0.7054687142372131,
      "learning_rate": 2.466429710702893e-07,
      "loss": 2.1627,
      "step": 514
    },
    {
      "epoch": 0.1256404000975848,
      "grad_norm": 1.6336970329284668,
      "learning_rate": 2.466356072067116e-07,
      "loss": 1.9868,
      "step": 515
    },
    {
      "epoch": 0.12588436203952183,
      "grad_norm": 0.5906395316123962,
      "learning_rate": 2.4662823969625266e-07,
      "loss": 2.0956,
      "step": 516
    },
    {
      "epoch": 0.1261283239814589,
      "grad_norm": 0.6716820001602173,
      "learning_rate": 2.466208685362027e-07,
      "loss": 1.7824,
      "step": 517
    },
    {
      "epoch": 0.12637228592339594,
      "grad_norm": 0.6512242555618286,
      "learning_rate": 2.4661349372384934e-07,
      "loss": 1.9613,
      "step": 518
    },
    {
      "epoch": 0.12661624786533301,
      "grad_norm": 0.6093871593475342,
      "learning_rate": 2.466061152564773e-07,
      "loss": 2.133,
      "step": 519
    },
    {
      "epoch": 0.12686020980727006,
      "grad_norm": 0.6348795890808105,
      "learning_rate": 2.4659873313136873e-07,
      "loss": 1.9917,
      "step": 520
    },
    {
      "epoch": 0.12686020980727006,
      "eval_loss": 2.032553195953369,
      "eval_runtime": 82.4844,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 520
    },
    {
      "epoch": 0.12710417174920713,
      "grad_norm": 0.7311941385269165,
      "learning_rate": 2.465913473458031e-07,
      "loss": 1.989,
      "step": 521
    },
    {
      "epoch": 0.12734813369114417,
      "grad_norm": 0.6399905681610107,
      "learning_rate": 2.465839578970572e-07,
      "loss": 2.0483,
      "step": 522
    },
    {
      "epoch": 0.12759209563308124,
      "grad_norm": 0.6352974772453308,
      "learning_rate": 2.46576564782405e-07,
      "loss": 2.1475,
      "step": 523
    },
    {
      "epoch": 0.12783605757501829,
      "grad_norm": 1.216728687286377,
      "learning_rate": 2.4656916799911783e-07,
      "loss": 2.1861,
      "step": 524
    },
    {
      "epoch": 0.12808001951695536,
      "grad_norm": 16.101139068603516,
      "learning_rate": 2.4656176754446437e-07,
      "loss": 2.1574,
      "step": 525
    },
    {
      "epoch": 0.1283239814588924,
      "grad_norm": 0.6247526407241821,
      "learning_rate": 2.4655436341571053e-07,
      "loss": 2.1521,
      "step": 526
    },
    {
      "epoch": 0.12856794340082947,
      "grad_norm": 0.6813322305679321,
      "learning_rate": 2.4654695561011943e-07,
      "loss": 2.1452,
      "step": 527
    },
    {
      "epoch": 0.12881190534276654,
      "grad_norm": 1.005342721939087,
      "learning_rate": 2.4653954412495173e-07,
      "loss": 2.0553,
      "step": 528
    },
    {
      "epoch": 0.12905586728470358,
      "grad_norm": 0.6594173312187195,
      "learning_rate": 2.46532128957465e-07,
      "loss": 2.0046,
      "step": 529
    },
    {
      "epoch": 0.12929982922664066,
      "grad_norm": 0.6536688208580017,
      "learning_rate": 2.465247101049144e-07,
      "loss": 2.0084,
      "step": 530
    },
    {
      "epoch": 0.12929982922664066,
      "eval_loss": 2.0311105251312256,
      "eval_runtime": 82.4257,
      "eval_samples_per_second": 3.106,
      "eval_steps_per_second": 0.776,
      "step": 530
    },
    {
      "epoch": 0.1295437911685777,
      "grad_norm": 0.6012054085731506,
      "learning_rate": 2.465172875645522e-07,
      "loss": 2.2109,
      "step": 531
    },
    {
      "epoch": 0.12978775311051477,
      "grad_norm": 0.7324656248092651,
      "learning_rate": 2.4650986133362793e-07,
      "loss": 2.0499,
      "step": 532
    },
    {
      "epoch": 0.1300317150524518,
      "grad_norm": 0.6712090969085693,
      "learning_rate": 2.465024314093885e-07,
      "loss": 1.9613,
      "step": 533
    },
    {
      "epoch": 0.13027567699438888,
      "grad_norm": 0.8287779688835144,
      "learning_rate": 2.4649499778907805e-07,
      "loss": 2.0791,
      "step": 534
    },
    {
      "epoch": 0.13051963893632593,
      "grad_norm": 0.7180325388908386,
      "learning_rate": 2.4648756046993777e-07,
      "loss": 2.1055,
      "step": 535
    },
    {
      "epoch": 0.130763600878263,
      "grad_norm": 0.8076268434524536,
      "learning_rate": 2.4648011944920643e-07,
      "loss": 2.0083,
      "step": 536
    },
    {
      "epoch": 0.13100756282020004,
      "grad_norm": 0.6361818909645081,
      "learning_rate": 2.464726747241198e-07,
      "loss": 2.069,
      "step": 537
    },
    {
      "epoch": 0.1312515247621371,
      "grad_norm": 0.6086621284484863,
      "learning_rate": 2.46465226291911e-07,
      "loss": 1.9604,
      "step": 538
    },
    {
      "epoch": 0.13149548670407415,
      "grad_norm": 0.5492677092552185,
      "learning_rate": 2.4645777414981045e-07,
      "loss": 2.0372,
      "step": 539
    },
    {
      "epoch": 0.13173944864601123,
      "grad_norm": 0.7199727892875671,
      "learning_rate": 2.4645031829504564e-07,
      "loss": 2.0154,
      "step": 540
    },
    {
      "epoch": 0.13173944864601123,
      "eval_loss": 2.030088186264038,
      "eval_runtime": 82.3631,
      "eval_samples_per_second": 3.108,
      "eval_steps_per_second": 0.777,
      "step": 540
    },
    {
      "epoch": 0.13198341058794827,
      "grad_norm": 1.0681871175765991,
      "learning_rate": 2.464428587248415e-07,
      "loss": 2.1332,
      "step": 541
    },
    {
      "epoch": 0.13222737252988534,
      "grad_norm": 2.4582200050354004,
      "learning_rate": 2.4643539543642e-07,
      "loss": 2.0242,
      "step": 542
    },
    {
      "epoch": 0.13247133447182238,
      "grad_norm": 0.556507408618927,
      "learning_rate": 2.4642792842700055e-07,
      "loss": 2.0096,
      "step": 543
    },
    {
      "epoch": 0.13271529641375945,
      "grad_norm": 0.6344817280769348,
      "learning_rate": 2.464204576937995e-07,
      "loss": 1.9323,
      "step": 544
    },
    {
      "epoch": 0.13295925835569652,
      "grad_norm": 0.6195700764656067,
      "learning_rate": 2.4641298323403077e-07,
      "loss": 2.1069,
      "step": 545
    },
    {
      "epoch": 0.13320322029763357,
      "grad_norm": 0.7518613338470459,
      "learning_rate": 2.464055050449052e-07,
      "loss": 2.0032,
      "step": 546
    },
    {
      "epoch": 0.13344718223957064,
      "grad_norm": 0.710758626461029,
      "learning_rate": 2.46398023123631e-07,
      "loss": 2.1871,
      "step": 547
    },
    {
      "epoch": 0.13369114418150768,
      "grad_norm": 0.6719003915786743,
      "learning_rate": 2.463905374674136e-07,
      "loss": 1.9962,
      "step": 548
    },
    {
      "epoch": 0.13393510612344475,
      "grad_norm": 0.9001761078834534,
      "learning_rate": 2.4638304807345555e-07,
      "loss": 2.0175,
      "step": 549
    },
    {
      "epoch": 0.1341790680653818,
      "grad_norm": 0.8284860849380493,
      "learning_rate": 2.463755549389567e-07,
      "loss": 2.0108,
      "step": 550
    },
    {
      "epoch": 0.1341790680653818,
      "eval_loss": 2.0290093421936035,
      "eval_runtime": 82.3581,
      "eval_samples_per_second": 3.108,
      "eval_steps_per_second": 0.777,
      "step": 550
    },
    {
      "epoch": 0.13442303000731887,
      "grad_norm": 0.5253940224647522,
      "learning_rate": 2.46368058061114e-07,
      "loss": 2.055,
      "step": 551
    },
    {
      "epoch": 0.1346669919492559,
      "grad_norm": 0.7425602078437805,
      "learning_rate": 2.4636055743712173e-07,
      "loss": 2.1348,
      "step": 552
    },
    {
      "epoch": 0.13491095389119298,
      "grad_norm": 0.5845130681991577,
      "learning_rate": 2.4635305306417126e-07,
      "loss": 2.0106,
      "step": 553
    },
    {
      "epoch": 0.13515491583313002,
      "grad_norm": 0.6662041544914246,
      "learning_rate": 2.463455449394512e-07,
      "loss": 2.1305,
      "step": 554
    },
    {
      "epoch": 0.1353988777750671,
      "grad_norm": 0.6484659910202026,
      "learning_rate": 2.4633803306014726e-07,
      "loss": 2.1217,
      "step": 555
    },
    {
      "epoch": 0.13564283971700414,
      "grad_norm": 0.7985727787017822,
      "learning_rate": 2.4633051742344244e-07,
      "loss": 2.1338,
      "step": 556
    },
    {
      "epoch": 0.1358868016589412,
      "grad_norm": 0.6523579359054565,
      "learning_rate": 2.463229980265169e-07,
      "loss": 2.0912,
      "step": 557
    },
    {
      "epoch": 0.13613076360087825,
      "grad_norm": 0.7691523432731628,
      "learning_rate": 2.4631547486654805e-07,
      "loss": 1.9642,
      "step": 558
    },
    {
      "epoch": 0.13637472554281532,
      "grad_norm": 0.7172589898109436,
      "learning_rate": 2.4630794794071024e-07,
      "loss": 1.974,
      "step": 559
    },
    {
      "epoch": 0.13661868748475237,
      "grad_norm": 0.9193669557571411,
      "learning_rate": 2.4630041724617526e-07,
      "loss": 2.0944,
      "step": 560
    },
    {
      "epoch": 0.13661868748475237,
      "eval_loss": 2.027602195739746,
      "eval_runtime": 82.4834,
      "eval_samples_per_second": 3.104,
      "eval_steps_per_second": 0.776,
      "step": 560
    }
  ],
  "logging_steps": 1,
  "max_steps": 4099,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.767317518548992e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}