| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.791738382099828, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.99965514863094e-07, | |
| "loss": 0.0333, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 9.99862059452376e-07, | |
| "loss": 0.0332, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 9.99758604041658e-07, | |
| "loss": 0.0336, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 0.21229705214500427, | |
| "eval_runtime": 487.1939, | |
| "eval_samples_per_second": 10.649, | |
| "eval_steps_per_second": 1.332, | |
| "eval_wer": 0.1448122092799754, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 9.9965514863094e-07, | |
| "loss": 0.0329, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 9.995516932202221e-07, | |
| "loss": 0.0323, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 9.99448237809504e-07, | |
| "loss": 0.0317, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 0.20801204442977905, | |
| "eval_runtime": 282.77, | |
| "eval_samples_per_second": 18.347, | |
| "eval_steps_per_second": 2.295, | |
| "eval_wer": 0.14550417099142737, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 9.993454721015242e-07, | |
| "loss": 0.0323, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 9.992420166908062e-07, | |
| "loss": 0.031, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 9.991385612800883e-07, | |
| "loss": 0.0314, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 9.990351058693703e-07, | |
| "loss": 0.0317, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 0.2071654200553894, | |
| "eval_runtime": 303.1157, | |
| "eval_samples_per_second": 17.116, | |
| "eval_steps_per_second": 2.141, | |
| "eval_wer": 0.14465844001076386, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 9.989316504586522e-07, | |
| "loss": 0.031, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 9.988281950479342e-07, | |
| "loss": 0.0318, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 9.987247396372162e-07, | |
| "loss": 0.0314, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 0.20706616342067719, | |
| "eval_runtime": 287.3918, | |
| "eval_samples_per_second": 18.052, | |
| "eval_steps_per_second": 2.258, | |
| "eval_wer": 0.1442355745204321, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 1450000, | |
| "num_train_epochs": 10000, | |
| "total_flos": 5.652539665735567e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |