| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7127355796687607, | |
| "eval_steps": 13, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.018275271273557967, | |
| "grad_norm": 4.217789649963379, | |
| "learning_rate": 5e-05, | |
| "loss": 8.8561, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.018275271273557967, | |
| "eval_loss": 8.832687377929688, | |
| "eval_runtime": 66.6285, | |
| "eval_samples_per_second": 5.538, | |
| "eval_steps_per_second": 1.396, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.036550542547115934, | |
| "grad_norm": 4.4302568435668945, | |
| "learning_rate": 0.0001, | |
| "loss": 8.7806, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0548258138206739, | |
| "grad_norm": 4.538593292236328, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 8.662, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07310108509423187, | |
| "grad_norm": 5.8020195960998535, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 7.9472, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.09137635636778983, | |
| "grad_norm": 6.238458633422852, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 6.4558, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1096516276413478, | |
| "grad_norm": 5.397599220275879, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 4.8986, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.12792689891490577, | |
| "grad_norm": 4.881111145019531, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 4.1723, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.14620217018846374, | |
| "grad_norm": 3.493053436279297, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.2577, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1644774414620217, | |
| "grad_norm": 2.9015204906463623, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 2.7507, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.18275271273557966, | |
| "grad_norm": 2.8159022331237793, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.5523, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20102798400913763, | |
| "grad_norm": 3.2096753120422363, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 2.0235, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2193032552826956, | |
| "grad_norm": 3.640693426132202, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 1.4851, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.23757852655625358, | |
| "grad_norm": 3.8002378940582275, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 1.0231, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.23757852655625358, | |
| "eval_loss": 0.6289411783218384, | |
| "eval_runtime": 66.1722, | |
| "eval_samples_per_second": 5.576, | |
| "eval_steps_per_second": 1.405, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.25585379782981155, | |
| "grad_norm": 2.663853168487549, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.6116, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.2741290691033695, | |
| "grad_norm": 2.392077922821045, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 0.3944, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2924043403769275, | |
| "grad_norm": 2.333585262298584, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 0.2627, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3106796116504854, | |
| "grad_norm": 0.5413778424263, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1026, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.3289548829240434, | |
| "grad_norm": 1.2651410102844238, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.2252, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3472301541976014, | |
| "grad_norm": 0.9693997502326965, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 0.0994, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.36550542547115933, | |
| "grad_norm": 1.002900242805481, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1058, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3837806967447173, | |
| "grad_norm": 0.8303418755531311, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 0.0462, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.40205596801827526, | |
| "grad_norm": 0.6390141248703003, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.0405, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.42033123929183325, | |
| "grad_norm": 0.6512762308120728, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0238, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.4386065105653912, | |
| "grad_norm": 0.7656252980232239, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.0375, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.4568817818389492, | |
| "grad_norm": 0.46867817640304565, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 0.016, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.47515705311250717, | |
| "grad_norm": 0.22703154385089874, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0108, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.47515705311250717, | |
| "eval_loss": 0.011774315498769283, | |
| "eval_runtime": 66.1555, | |
| "eval_samples_per_second": 5.578, | |
| "eval_steps_per_second": 1.406, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.4934323243860651, | |
| "grad_norm": 0.2927595376968384, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 0.0162, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5117075956596231, | |
| "grad_norm": 0.1291339248418808, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.0014, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.5299828669331811, | |
| "grad_norm": 0.24390743672847748, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.01, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.548258138206739, | |
| "grad_norm": 0.07123029232025146, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.0007, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.566533409480297, | |
| "grad_norm": 0.03508315980434418, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 0.0004, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.584808680753855, | |
| "grad_norm": 0.33584150671958923, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0116, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.6030839520274129, | |
| "grad_norm": 0.18450838327407837, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 0.0017, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6213592233009708, | |
| "grad_norm": 0.30941474437713623, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0089, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.6396344945745288, | |
| "grad_norm": 0.3536520004272461, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0024, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.6579097658480868, | |
| "grad_norm": 0.3076607882976532, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.0032, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6761850371216448, | |
| "grad_norm": 0.09125807881355286, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 0.005, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.6944603083952028, | |
| "grad_norm": 0.09541966021060944, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0006, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.7127355796687607, | |
| "grad_norm": 0.042761899530887604, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 0.0003, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7127355796687607, | |
| "eval_loss": 0.005322535987943411, | |
| "eval_runtime": 66.1258, | |
| "eval_samples_per_second": 5.58, | |
| "eval_steps_per_second": 1.406, | |
| "step": 39 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5844083502531215e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |