| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9939659901261657, | |
| "eval_steps": 500, | |
| "global_step": 302, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006582556226001097, | |
| "grad_norm": 2.453043920718015, | |
| "learning_rate": 3.2258064516129035e-07, | |
| "loss": 2.2482, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03291278113000549, | |
| "grad_norm": 2.0003115811727206, | |
| "learning_rate": 1.6129032258064516e-06, | |
| "loss": 2.0687, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06582556226001098, | |
| "grad_norm": 0.9576535879971833, | |
| "learning_rate": 3.225806451612903e-06, | |
| "loss": 1.6905, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09873834339001646, | |
| "grad_norm": 0.30116763306370925, | |
| "learning_rate": 4.838709677419355e-06, | |
| "loss": 1.0142, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.13165112452002195, | |
| "grad_norm": 0.20223223062062798, | |
| "learning_rate": 6.451612903225806e-06, | |
| "loss": 0.7111, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16456390565002743, | |
| "grad_norm": 0.14657727394091244, | |
| "learning_rate": 8.064516129032258e-06, | |
| "loss": 0.6454, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1974766867800329, | |
| "grad_norm": 0.11208122922317201, | |
| "learning_rate": 9.67741935483871e-06, | |
| "loss": 0.5687, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2303894679100384, | |
| "grad_norm": 0.1073134953639841, | |
| "learning_rate": 9.99462543481167e-06, | |
| "loss": 0.5147, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2633022490400439, | |
| "grad_norm": 0.13688522912038725, | |
| "learning_rate": 9.9728110648995e-06, | |
| "loss": 0.4698, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.29621503017004935, | |
| "grad_norm": 0.0918265619598427, | |
| "learning_rate": 9.934294192840518e-06, | |
| "loss": 0.4815, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.32912781130005486, | |
| "grad_norm": 0.11050070883185271, | |
| "learning_rate": 9.879204187744036e-06, | |
| "loss": 0.453, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3620405924300603, | |
| "grad_norm": 0.07394441588683744, | |
| "learning_rate": 9.807726083973192e-06, | |
| "loss": 0.4228, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.3949533735600658, | |
| "grad_norm": 0.09321757955302397, | |
| "learning_rate": 9.720099959658062e-06, | |
| "loss": 0.4392, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.42786615469007133, | |
| "grad_norm": 0.09437542122654453, | |
| "learning_rate": 9.61662013032972e-06, | |
| "loss": 0.4304, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.4607789358200768, | |
| "grad_norm": 0.07578178267682123, | |
| "learning_rate": 9.497634160383627e-06, | |
| "loss": 0.4391, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4936917169500823, | |
| "grad_norm": 0.07062146422446726, | |
| "learning_rate": 9.36354169569261e-06, | |
| "loss": 0.4149, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5266044980800878, | |
| "grad_norm": 0.07563993059966992, | |
| "learning_rate": 9.214793121290442e-06, | |
| "loss": 0.424, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5595172792100932, | |
| "grad_norm": 0.0714902674482327, | |
| "learning_rate": 9.051888048634471e-06, | |
| "loss": 0.4538, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.5924300603400987, | |
| "grad_norm": 0.06635393353384471, | |
| "learning_rate": 8.875373637528336e-06, | |
| "loss": 0.4058, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6253428414701042, | |
| "grad_norm": 0.06813778711254845, | |
| "learning_rate": 8.685842758340912e-06, | |
| "loss": 0.3843, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.6582556226001097, | |
| "grad_norm": 0.06038001684085916, | |
| "learning_rate": 8.483932000694295e-06, | |
| "loss": 0.3839, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6911684037301152, | |
| "grad_norm": 0.06387985587791511, | |
| "learning_rate": 8.270319535309035e-06, | |
| "loss": 0.3949, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.7240811848601206, | |
| "grad_norm": 0.07413216886866929, | |
| "learning_rate": 8.04572283618829e-06, | |
| "loss": 0.3857, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7569939659901261, | |
| "grad_norm": 0.06197794577102131, | |
| "learning_rate": 7.810896270791484e-06, | |
| "loss": 0.3867, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.7899067471201316, | |
| "grad_norm": 0.06775910453909832, | |
| "learning_rate": 7.566628566291537e-06, | |
| "loss": 0.4253, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8228195282501372, | |
| "grad_norm": 0.06335190737071104, | |
| "learning_rate": 7.313740160425887e-06, | |
| "loss": 0.3592, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.8557323093801427, | |
| "grad_norm": 0.06053821259981546, | |
| "learning_rate": 7.053080445839211e-06, | |
| "loss": 0.3982, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8886450905101481, | |
| "grad_norm": 0.05817539529457328, | |
| "learning_rate": 6.7855249171734e-06, | |
| "loss": 0.3825, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.9215578716401536, | |
| "grad_norm": 0.06281193554143008, | |
| "learning_rate": 6.511972230487091e-06, | |
| "loss": 0.3455, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9544706527701591, | |
| "grad_norm": 0.057663764408870714, | |
| "learning_rate": 6.2333411848814415e-06, | |
| "loss": 0.3971, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.9873834339001646, | |
| "grad_norm": 0.06667224024941, | |
| "learning_rate": 5.95056763647016e-06, | |
| "loss": 0.3715, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9939659901261657, | |
| "eval_loss": 0.40147778391838074, | |
| "eval_runtime": 33.4304, | |
| "eval_samples_per_second": 20.61, | |
| "eval_steps_per_second": 5.175, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 1.0263302249040045, | |
| "grad_norm": 0.06761665079835917, | |
| "learning_rate": 5.664601355059044e-06, | |
| "loss": 0.4201, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.05924300603401, | |
| "grad_norm": 0.06719806083275706, | |
| "learning_rate": 5.376402834092721e-06, | |
| "loss": 0.3496, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0921557871640153, | |
| "grad_norm": 0.05895174295699646, | |
| "learning_rate": 5.086940064583179e-06, | |
| "loss": 0.3448, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.125068568294021, | |
| "grad_norm": 0.05113621376287118, | |
| "learning_rate": 4.797185283855756e-06, | |
| "loss": 0.3397, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1579813494240263, | |
| "grad_norm": 0.0675921905296945, | |
| "learning_rate": 4.5081117100327594e-06, | |
| "loss": 0.3611, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.1908941305540317, | |
| "grad_norm": 0.05801392979206245, | |
| "learning_rate": 4.220690273222802e-06, | |
| "loss": 0.3465, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.2238069116840373, | |
| "grad_norm": 0.07011122057291769, | |
| "learning_rate": 3.935886354395057e-06, | |
| "loss": 0.3342, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.2567196928140427, | |
| "grad_norm": 0.06127719545378434, | |
| "learning_rate": 3.6546565428917623e-06, | |
| "loss": 0.353, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.2896324739440483, | |
| "grad_norm": 0.05597451127292148, | |
| "learning_rate": 3.377945423469727e-06, | |
| "loss": 0.3267, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.3225452550740537, | |
| "grad_norm": 0.06363474070842762, | |
| "learning_rate": 3.1066824036624086e-06, | |
| "loss": 0.3667, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3554580362040594, | |
| "grad_norm": 0.06166154027576164, | |
| "learning_rate": 2.84177859211873e-06, | |
| "loss": 0.355, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.3883708173340648, | |
| "grad_norm": 0.06396509045884582, | |
| "learning_rate": 2.584123738403527e-06, | |
| "loss": 0.3663, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4212835984640702, | |
| "grad_norm": 0.060133414345748225, | |
| "learning_rate": 2.3345832445381415e-06, | |
| "loss": 0.3389, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.4541963795940758, | |
| "grad_norm": 0.05614627720977617, | |
| "learning_rate": 2.0939952583186806e-06, | |
| "loss": 0.3233, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.4871091607240812, | |
| "grad_norm": 0.06341047469554213, | |
| "learning_rate": 1.8631678581748059e-06, | |
| "loss": 0.3828, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5200219418540866, | |
| "grad_norm": 0.0586104014108502, | |
| "learning_rate": 1.6428763390244462e-06, | |
| "loss": 0.3142, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5529347229840922, | |
| "grad_norm": 0.06293576847799762, | |
| "learning_rate": 1.4338606082406232e-06, | |
| "loss": 0.3617, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.5858475041140978, | |
| "grad_norm": 0.05673087625262431, | |
| "learning_rate": 1.236822700476683e-06, | |
| "loss": 0.3574, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.618760285244103, | |
| "grad_norm": 0.06411343693892131, | |
| "learning_rate": 1.0524244196971228e-06, | |
| "loss": 0.3708, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.6516730663741086, | |
| "grad_norm": 0.05756395567753849, | |
| "learning_rate": 8.812851163337976e-07, | |
| "loss": 0.3201, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6845858475041142, | |
| "grad_norm": 0.06491241755900852, | |
| "learning_rate": 7.239796070335775e-07, | |
| "loss": 0.3551, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.7174986286341196, | |
| "grad_norm": 0.06269841166969671, | |
| "learning_rate": 5.810362439844897e-07, | |
| "loss": 0.3198, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.750411409764125, | |
| "grad_norm": 0.06872125923905117, | |
| "learning_rate": 4.529351403050586e-07, | |
| "loss": 0.3461, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.7833241908941306, | |
| "grad_norm": 0.07301293362304442, | |
| "learning_rate": 3.401065574573187e-07, | |
| "loss": 0.389, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.816236972024136, | |
| "grad_norm": 0.06735671030896039, | |
| "learning_rate": 2.4292946009982163e-07, | |
| "loss": 0.3653, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.8491497531541414, | |
| "grad_norm": 0.05540343390769577, | |
| "learning_rate": 1.6173024323451747e-07, | |
| "loss": 0.341, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.882062534284147, | |
| "grad_norm": 0.05948050374671173, | |
| "learning_rate": 9.678163592275858e-08, | |
| "loss": 0.3237, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.9149753154141524, | |
| "grad_norm": 0.05494271383448326, | |
| "learning_rate": 4.830178525256079e-08, | |
| "loss": 0.3365, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.9478880965441578, | |
| "grad_norm": 0.052912085010542946, | |
| "learning_rate": 1.6453523633868096e-08, | |
| "loss": 0.301, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.9808008776741635, | |
| "grad_norm": 0.0646409275755533, | |
| "learning_rate": 1.3438218828076831e-09, | |
| "loss": 0.3628, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.9939659901261657, | |
| "eval_loss": 0.39538663625717163, | |
| "eval_runtime": 33.2149, | |
| "eval_samples_per_second": 20.744, | |
| "eval_steps_per_second": 5.209, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 1.9939659901261657, | |
| "step": 302, | |
| "total_flos": 6.708841195629445e+17, | |
| "train_loss": 0.4515860276111704, | |
| "train_runtime": 2865.436, | |
| "train_samples_per_second": 5.089, | |
| "train_steps_per_second": 0.105 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 302, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.708841195629445e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |