| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.407407407407407, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0625, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": 0.28051668405532837, | |
| "eval_runtime": 198.6071, | |
| "eval_samples_per_second": 16.374, | |
| "eval_steps_per_second": 2.049, | |
| "eval_wer": 0.2381097005406062, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 9.987639060568604e-05, | |
| "loss": 0.047, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_loss": 0.34423157572746277, | |
| "eval_runtime": 147.7154, | |
| "eval_samples_per_second": 22.015, | |
| "eval_steps_per_second": 2.755, | |
| "eval_wer": 0.2334325457085586, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 9.975278121137207e-05, | |
| "loss": 0.0409, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 0.3597787916660309, | |
| "eval_runtime": 156.092, | |
| "eval_samples_per_second": 20.834, | |
| "eval_steps_per_second": 2.607, | |
| "eval_wer": 0.23124582396889995, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 9.96291718170581e-05, | |
| "loss": 0.0413, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_loss": 0.35860675573349, | |
| "eval_runtime": 151.5312, | |
| "eval_samples_per_second": 21.461, | |
| "eval_steps_per_second": 2.686, | |
| "eval_wer": 0.2421794326671931, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 9.950556242274414e-05, | |
| "loss": 0.0388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "eval_loss": 0.3748931586742401, | |
| "eval_runtime": 154.1563, | |
| "eval_samples_per_second": 21.095, | |
| "eval_steps_per_second": 2.64, | |
| "eval_wer": 0.23914231913988945, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 9.938195302843017e-05, | |
| "loss": 0.0383, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_loss": 0.36109668016433716, | |
| "eval_runtime": 163.2656, | |
| "eval_samples_per_second": 19.918, | |
| "eval_steps_per_second": 2.493, | |
| "eval_wer": 0.2352548138249408, | |
| "step": 600 | |
| } | |
| ], | |
| "max_steps": 81000, | |
| "num_train_epochs": 1000, | |
| "total_flos": 1.9144703855961313e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |