| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.906542056074766, | |
| "global_step": 3200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.000285, | |
| "loss": 3.7926, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "eval_loss": 2.7348811626434326, | |
| "eval_runtime": 77.58, | |
| "eval_samples_per_second": 5.955, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 0.0002584870848708487, | |
| "loss": 0.6512, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "eval_loss": 0.3463097810745239, | |
| "eval_runtime": 78.064, | |
| "eval_samples_per_second": 5.918, | |
| "eval_wer": 0.44352893890675243, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "learning_rate": 0.00021420664206642064, | |
| "loss": 0.2406, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "eval_loss": 0.2929766774177551, | |
| "eval_runtime": 77.4439, | |
| "eval_samples_per_second": 5.966, | |
| "eval_wer": 0.38183279742765275, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 0.0001699261992619926, | |
| "loss": 0.153, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "eval_loss": 0.29108402132987976, | |
| "eval_runtime": 77.5593, | |
| "eval_samples_per_second": 5.957, | |
| "eval_wer": 0.3659565916398714, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 0.00012564575645756455, | |
| "loss": 0.1189, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "eval_loss": 0.3000461161136627, | |
| "eval_runtime": 77.8803, | |
| "eval_samples_per_second": 5.932, | |
| "eval_wer": 0.3516881028938907, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 22.43, | |
| "learning_rate": 8.136531365313652e-05, | |
| "loss": 0.0902, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 22.43, | |
| "eval_loss": 0.31765106320381165, | |
| "eval_runtime": 77.559, | |
| "eval_samples_per_second": 5.957, | |
| "eval_wer": 0.3432475884244373, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 26.17, | |
| "learning_rate": 3.7084870848708486e-05, | |
| "loss": 0.0748, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 26.17, | |
| "eval_loss": 0.32380491495132446, | |
| "eval_runtime": 77.8712, | |
| "eval_samples_per_second": 5.933, | |
| "eval_wer": 0.33641479099678456, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 29.91, | |
| "learning_rate": 0.0, | |
| "loss": 0.0659, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 29.91, | |
| "eval_loss": 0.3231419026851654, | |
| "eval_runtime": 77.4758, | |
| "eval_samples_per_second": 5.963, | |
| "eval_wer": 0.3307877813504823, | |
| "step": 3200 | |
| } | |
| ], | |
| "max_steps": 3210, | |
| "num_train_epochs": 30, | |
| "total_flos": 2.032238891438037e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |