| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 28.11244979919679, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.97e-05, | |
| "loss": 4.8083, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 2.9298839569091797, | |
| "eval_runtime": 72.494, | |
| "eval_samples_per_second": 23.174, | |
| "eval_steps_per_second": 2.897, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 1.5432, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_loss": 0.3672347962856293, | |
| "eval_runtime": 72.6811, | |
| "eval_samples_per_second": 23.115, | |
| "eval_steps_per_second": 2.889, | |
| "eval_wer": 0.41520226035421404, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 9.231839258114375e-05, | |
| "loss": 0.305, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "eval_loss": 0.2641327381134033, | |
| "eval_runtime": 74.7155, | |
| "eval_samples_per_second": 22.485, | |
| "eval_steps_per_second": 2.811, | |
| "eval_wer": 0.32127351664254705, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 8.459041731066461e-05, | |
| "loss": 0.1614, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "eval_loss": 0.2683059871196747, | |
| "eval_runtime": 73.5962, | |
| "eval_samples_per_second": 22.827, | |
| "eval_steps_per_second": 2.853, | |
| "eval_wer": 0.2949486596375164, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 7.686244204018547e-05, | |
| "loss": 0.1096, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "eval_loss": 0.25842466950416565, | |
| "eval_runtime": 72.9953, | |
| "eval_samples_per_second": 23.015, | |
| "eval_steps_per_second": 2.877, | |
| "eval_wer": 0.2761353455998897, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 6.913446676970633e-05, | |
| "loss": 0.0849, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "eval_loss": 0.2633412778377533, | |
| "eval_runtime": 73.117, | |
| "eval_samples_per_second": 22.977, | |
| "eval_steps_per_second": 2.872, | |
| "eval_wer": 0.26834814967955345, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 6.14064914992272e-05, | |
| "loss": 0.0666, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "eval_loss": 0.2584041357040405, | |
| "eval_runtime": 72.8822, | |
| "eval_samples_per_second": 23.051, | |
| "eval_steps_per_second": 2.881, | |
| "eval_wer": 0.2597339949004204, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "learning_rate": 5.3678516228748066e-05, | |
| "loss": 0.0558, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "eval_loss": 0.25760215520858765, | |
| "eval_runtime": 72.4376, | |
| "eval_samples_per_second": 23.192, | |
| "eval_steps_per_second": 2.899, | |
| "eval_wer": 0.24939700916546068, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 4.5950540958268935e-05, | |
| "loss": 0.0455, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "eval_loss": 0.2717011272907257, | |
| "eval_runtime": 73.3321, | |
| "eval_samples_per_second": 22.909, | |
| "eval_steps_per_second": 2.864, | |
| "eval_wer": 0.2531183240300462, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 20.08, | |
| "learning_rate": 3.8222565687789804e-05, | |
| "loss": 0.0422, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 20.08, | |
| "eval_loss": 0.29290544986724854, | |
| "eval_runtime": 74.3543, | |
| "eval_samples_per_second": 22.595, | |
| "eval_steps_per_second": 2.824, | |
| "eval_wer": 0.2513265798359865, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 22.09, | |
| "learning_rate": 3.0494590417310663e-05, | |
| "loss": 0.0383, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 22.09, | |
| "eval_loss": 0.27808326482772827, | |
| "eval_runtime": 73.2045, | |
| "eval_samples_per_second": 22.949, | |
| "eval_steps_per_second": 2.869, | |
| "eval_wer": 0.2480187444007994, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 24.1, | |
| "learning_rate": 2.2766615146831533e-05, | |
| "loss": 0.033, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 24.1, | |
| "eval_loss": 0.2956600785255432, | |
| "eval_runtime": 73.3873, | |
| "eval_samples_per_second": 22.892, | |
| "eval_steps_per_second": 2.862, | |
| "eval_wer": 0.2503617945007236, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 26.1, | |
| "learning_rate": 1.5038639876352395e-05, | |
| "loss": 0.0314, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 26.1, | |
| "eval_loss": 0.2865420877933502, | |
| "eval_runtime": 73.2994, | |
| "eval_samples_per_second": 22.92, | |
| "eval_steps_per_second": 2.865, | |
| "eval_wer": 0.24863896354489698, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 28.11, | |
| "learning_rate": 7.310664605873262e-06, | |
| "loss": 0.0308, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 28.11, | |
| "eval_loss": 0.29938989877700806, | |
| "eval_runtime": 73.3049, | |
| "eval_samples_per_second": 22.918, | |
| "eval_steps_per_second": 2.865, | |
| "eval_wer": 0.24732961201846876, | |
| "step": 7000 | |
| } | |
| ], | |
| "max_steps": 7470, | |
| "num_train_epochs": 30, | |
| "total_flos": 9.6467224304247e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |