| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 24.137931034482758, | |
| "global_step": 3500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.97e-05, | |
| "loss": 3.8466, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 2.4666473865509033, | |
| "eval_runtime": 66.0325, | |
| "eval_samples_per_second": 25.442, | |
| "eval_steps_per_second": 3.18, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 0.9403, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 0.4411664605140686, | |
| "eval_runtime": 66.1692, | |
| "eval_samples_per_second": 25.389, | |
| "eval_steps_per_second": 3.174, | |
| "eval_wer": 0.39928330232237613, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 8.516417910447761e-05, | |
| "loss": 0.2824, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 0.3629843592643738, | |
| "eval_runtime": 65.8604, | |
| "eval_samples_per_second": 25.508, | |
| "eval_steps_per_second": 3.189, | |
| "eval_wer": 0.3318172420922059, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 7.023880597014926e-05, | |
| "loss": 0.1639, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 0.3536781072616577, | |
| "eval_runtime": 65.6765, | |
| "eval_samples_per_second": 25.58, | |
| "eval_steps_per_second": 3.197, | |
| "eval_wer": 0.3030115085107849, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 5.53134328358209e-05, | |
| "loss": 0.1251, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "eval_loss": 0.39447325468063354, | |
| "eval_runtime": 66.1834, | |
| "eval_samples_per_second": 25.384, | |
| "eval_steps_per_second": 3.173, | |
| "eval_wer": 0.2973606229756736, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.69, | |
| "learning_rate": 4.0388059701492536e-05, | |
| "loss": 0.1075, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 20.69, | |
| "eval_loss": 0.3765473961830139, | |
| "eval_runtime": 66.4767, | |
| "eval_samples_per_second": 25.272, | |
| "eval_steps_per_second": 3.159, | |
| "eval_wer": 0.29370822134932123, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.14, | |
| "learning_rate": 2.546268656716418e-05, | |
| "loss": 0.0831, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 24.14, | |
| "eval_loss": 0.3624289631843567, | |
| "eval_runtime": 66.6948, | |
| "eval_samples_per_second": 25.189, | |
| "eval_steps_per_second": 3.149, | |
| "eval_wer": 0.2816484046585349, | |
| "step": 3500 | |
| } | |
| ], | |
| "max_steps": 4350, | |
| "num_train_epochs": 30, | |
| "total_flos": 3.2107549243060224e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |