| { | |
| "best_metric": 0.17608685300699967, | |
| "best_model_checkpoint": "train/2022-03-31-21-50-25/checkpoint-12960", | |
| "epoch": 1.9789280806229959, | |
| "global_step": 12960, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1209, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.1341254562139511, | |
| "eval_runtime": 468.0255, | |
| "eval_samples_per_second": 33.765, | |
| "eval_steps_per_second": 0.705, | |
| "eval_wer": 0.19390346491436916, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.8202695956065902e-05, | |
| "loss": 0.1034, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 0.13459239900112152, | |
| "eval_runtime": 466.4831, | |
| "eval_samples_per_second": 33.877, | |
| "eval_steps_per_second": 0.707, | |
| "eval_wer": 0.19213370791867054, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.6405391912131803e-05, | |
| "loss": 0.102, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 0.13314320147037506, | |
| "eval_runtime": 464.4247, | |
| "eval_samples_per_second": 34.027, | |
| "eval_steps_per_second": 0.711, | |
| "eval_wer": 0.1909591606748885, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.4609752038608755e-05, | |
| "loss": 0.1016, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 0.13013887405395508, | |
| "eval_runtime": 465.4732, | |
| "eval_samples_per_second": 33.95, | |
| "eval_steps_per_second": 0.709, | |
| "eval_wer": 0.1890465533387299, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.2812447994674656e-05, | |
| "loss": 0.109, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 0.12718452513217926, | |
| "eval_runtime": 465.6054, | |
| "eval_samples_per_second": 33.941, | |
| "eval_steps_per_second": 0.709, | |
| "eval_wer": 0.18734028538323572, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.1016808121151608e-05, | |
| "loss": 0.1251, | |
| "step": 6480 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 0.11952650547027588, | |
| "eval_runtime": 466.6012, | |
| "eval_samples_per_second": 33.868, | |
| "eval_steps_per_second": 0.707, | |
| "eval_wer": 0.1837848991317874, | |
| "step": 6480 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 9.219504077217509e-06, | |
| "loss": 0.118, | |
| "step": 7560 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 0.12100061029195786, | |
| "eval_runtime": 469.3607, | |
| "eval_samples_per_second": 33.669, | |
| "eval_steps_per_second": 0.703, | |
| "eval_wer": 0.18134057108391663, | |
| "step": 7560 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 7.422200033283409e-06, | |
| "loss": 0.1156, | |
| "step": 8640 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_loss": 0.12037315964698792, | |
| "eval_runtime": 463.2023, | |
| "eval_samples_per_second": 34.117, | |
| "eval_steps_per_second": 0.712, | |
| "eval_wer": 0.17957875021824357, | |
| "step": 8640 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 5.62656015976036e-06, | |
| "loss": 0.1152, | |
| "step": 9720 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.11911876499652863, | |
| "eval_runtime": 465.9842, | |
| "eval_samples_per_second": 33.913, | |
| "eval_steps_per_second": 0.708, | |
| "eval_wer": 0.17893592368617367, | |
| "step": 9720 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.829256115826261e-06, | |
| "loss": 0.1132, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_loss": 0.11926092952489853, | |
| "eval_runtime": 466.8163, | |
| "eval_samples_per_second": 33.853, | |
| "eval_steps_per_second": 0.707, | |
| "eval_wer": 0.1785629255749726, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.033616242303212e-06, | |
| "loss": 0.1133, | |
| "step": 11880 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_loss": 0.11755573004484177, | |
| "eval_runtime": 568.3919, | |
| "eval_samples_per_second": 27.803, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.17685665761947844, | |
| "step": 11880 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 2.36312198369113e-07, | |
| "loss": 0.111, | |
| "step": 12960 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 0.11704171448945999, | |
| "eval_runtime": 495.4724, | |
| "eval_samples_per_second": 31.895, | |
| "eval_steps_per_second": 0.666, | |
| "eval_wer": 0.17608685300699967, | |
| "step": 12960 | |
| } | |
| ], | |
| "max_steps": 13098, | |
| "num_train_epochs": 2, | |
| "total_flos": 5.0362098492856254e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |