| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08708430849616285, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001741686169923257, | |
| "grad_norm": 3.0921738147735596, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9628, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001741686169923257, | |
| "eval_loss": 2.522160053253174, | |
| "eval_runtime": 351.8165, | |
| "eval_samples_per_second": 10.994, | |
| "eval_steps_per_second": 2.749, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003483372339846514, | |
| "grad_norm": 2.935767412185669, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9583, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005225058509769771, | |
| "grad_norm": 2.823258399963379, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 1.834, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006966744679693028, | |
| "grad_norm": 2.366340160369873, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 1.4925, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.008708430849616285, | |
| "grad_norm": 1.9907324314117432, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 1.1095, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010450117019539542, | |
| "grad_norm": 1.629361867904663, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.7857, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.012191803189462799, | |
| "grad_norm": 1.6523230075836182, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 0.6975, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.013933489359386056, | |
| "grad_norm": 1.8730814456939697, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.5556, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01567517552930931, | |
| "grad_norm": 1.6527363061904907, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 0.3817, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01741686169923257, | |
| "grad_norm": 1.103572964668274, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.2157, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.019158547869155825, | |
| "grad_norm": 1.1308457851409912, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.2139, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.020900234039079084, | |
| "grad_norm": 1.191890001296997, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 0.1758, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02264192020900234, | |
| "grad_norm": 0.7013289928436279, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 0.149, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02264192020900234, | |
| "eval_loss": 0.14659395813941956, | |
| "eval_runtime": 352.1285, | |
| "eval_samples_per_second": 10.985, | |
| "eval_steps_per_second": 2.746, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.024383606378925598, | |
| "grad_norm": 0.8086413145065308, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.147, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.026125292548848853, | |
| "grad_norm": 0.9576364755630493, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 0.1463, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.027866978718772112, | |
| "grad_norm": 0.5651424527168274, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 0.1489, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.029608664888695367, | |
| "grad_norm": 0.7270596623420715, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1385, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.03135035105861862, | |
| "grad_norm": 0.7695695161819458, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.1344, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.033092037228541885, | |
| "grad_norm": 0.5415513515472412, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 0.1271, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03483372339846514, | |
| "grad_norm": 0.5811421871185303, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1157, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.036575409568388395, | |
| "grad_norm": 0.5054365992546082, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 0.1075, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03831709573831165, | |
| "grad_norm": 0.43094682693481445, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.0977, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.04005878190823491, | |
| "grad_norm": 0.35248297452926636, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0912, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.04180046807815817, | |
| "grad_norm": 0.46628761291503906, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.1272, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04354215424808142, | |
| "grad_norm": 0.37774521112442017, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 0.0911, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04528384041800468, | |
| "grad_norm": 0.318975567817688, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0788, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04528384041800468, | |
| "eval_loss": 0.08614670485258102, | |
| "eval_runtime": 352.1165, | |
| "eval_samples_per_second": 10.985, | |
| "eval_steps_per_second": 2.746, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04702552658792794, | |
| "grad_norm": 0.3819328546524048, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 0.0898, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.048767212757851196, | |
| "grad_norm": 0.38801172375679016, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.0728, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.05050889892777445, | |
| "grad_norm": 0.3896499276161194, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.0931, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.052250585097697706, | |
| "grad_norm": 0.44811469316482544, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.104, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05399227126762097, | |
| "grad_norm": 0.3136563003063202, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 0.071, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.055733957437544224, | |
| "grad_norm": 0.4052649736404419, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0825, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.05747564360746748, | |
| "grad_norm": 0.3598843216896057, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 0.0904, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.059217329777390734, | |
| "grad_norm": 0.3175484538078308, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0929, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.060959015947313996, | |
| "grad_norm": 0.3411295413970947, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0826, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06270070211723724, | |
| "grad_norm": 0.4295560419559479, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.0854, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0644423882871605, | |
| "grad_norm": 0.336063414812088, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 0.0659, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.06618407445708377, | |
| "grad_norm": 0.3456127643585205, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0928, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.06792576062700702, | |
| "grad_norm": 0.37239643931388855, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 0.0807, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06792576062700702, | |
| "eval_loss": 0.07401972264051437, | |
| "eval_runtime": 352.2856, | |
| "eval_samples_per_second": 10.98, | |
| "eval_steps_per_second": 2.745, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06966744679693028, | |
| "grad_norm": 0.39408788084983826, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.1016, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07140913296685354, | |
| "grad_norm": 0.3007913827896118, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0745, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.07315081913677679, | |
| "grad_norm": 0.33744704723358154, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.086, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07489250530670005, | |
| "grad_norm": 0.37013643980026245, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 0.0786, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0766341914766233, | |
| "grad_norm": 0.3822749853134155, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0882, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.07837587764654656, | |
| "grad_norm": 0.27670085430145264, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 0.0661, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08011756381646983, | |
| "grad_norm": 0.33175578713417053, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.0792, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.08185924998639307, | |
| "grad_norm": 0.2966095209121704, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0705, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.08360093615631634, | |
| "grad_norm": 0.3922402262687683, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 0.0891, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08534262232623958, | |
| "grad_norm": 0.40399035811424255, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 0.0725, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.08708430849616285, | |
| "grad_norm": 0.34264397621154785, | |
| "learning_rate": 0.0, | |
| "loss": 0.0832, | |
| "step": 50 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1870196536967168e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |