| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.0, | |
| "global_step": 2490, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 4.3036, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "eval_loss": 0.6421348452568054, | |
| "eval_runtime": 266.3964, | |
| "eval_samples_per_second": 6.922, | |
| "eval_wer": 0.7364348312511654, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 0.0002547738693467337, | |
| "loss": 0.2971, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "eval_loss": 0.39936190843582153, | |
| "eval_runtime": 260.3352, | |
| "eval_samples_per_second": 7.083, | |
| "eval_wer": 0.4973895207906023, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 14.46, | |
| "learning_rate": 0.0001944723618090452, | |
| "loss": 0.1098, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 14.46, | |
| "eval_loss": 0.4166050851345062, | |
| "eval_runtime": 265.7729, | |
| "eval_samples_per_second": 6.938, | |
| "eval_wer": 0.4447137796009696, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 19.28, | |
| "learning_rate": 0.00013417085427135678, | |
| "loss": 0.0664, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 19.28, | |
| "eval_loss": 0.4249325096607208, | |
| "eval_runtime": 265.3532, | |
| "eval_samples_per_second": 6.949, | |
| "eval_wer": 0.4381875815774753, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 24.1, | |
| "learning_rate": 7.386934673366834e-05, | |
| "loss": 0.0499, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 24.1, | |
| "eval_loss": 0.42595329880714417, | |
| "eval_runtime": 264.3813, | |
| "eval_samples_per_second": 6.975, | |
| "eval_wer": 0.42401640872645907, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "learning_rate": 1.35678391959799e-05, | |
| "loss": 0.0407, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "eval_loss": 0.43581095337867737, | |
| "eval_runtime": 272.7984, | |
| "eval_samples_per_second": 6.76, | |
| "eval_wer": 0.4190751445086705, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 2490, | |
| "total_flos": 1.3821944739107277e+19, | |
| "train_runtime": 26891.3209, | |
| "train_samples_per_second": 0.093 | |
| } | |
| ], | |
| "max_steps": 2490, | |
| "num_train_epochs": 30, | |
| "total_flos": 1.3821944739107277e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |