| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "global_step": 7500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.229895830154419, | |
| "eval_runtime": 183.9271, | |
| "eval_samples_per_second": 21.742, | |
| "eval_steps_per_second": 2.718, | |
| "eval_wer": 0.7600584866577312, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002982, | |
| "loss": 3.6201, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.49728861451148987, | |
| "eval_runtime": 182.3086, | |
| "eval_samples_per_second": 21.935, | |
| "eval_steps_per_second": 2.743, | |
| "eval_wer": 0.42488119897648347, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002787, | |
| "loss": 0.6369, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.3747561573982239, | |
| "eval_runtime": 183.2249, | |
| "eval_samples_per_second": 21.826, | |
| "eval_steps_per_second": 2.729, | |
| "eval_wer": 0.3480930912635555, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00025727142857142856, | |
| "loss": 0.455, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 0.28336068987846375, | |
| "eval_runtime": 182.904, | |
| "eval_samples_per_second": 21.864, | |
| "eval_steps_per_second": 2.734, | |
| "eval_wer": 0.2643840623857682, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.00023584285714285712, | |
| "loss": 0.3177, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 0.24264603853225708, | |
| "eval_runtime": 183.432, | |
| "eval_samples_per_second": 21.801, | |
| "eval_steps_per_second": 2.726, | |
| "eval_wer": 0.22339466309248202, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.18675319850444794, | |
| "eval_runtime": 183.8214, | |
| "eval_samples_per_second": 21.755, | |
| "eval_steps_per_second": 2.72, | |
| "eval_wer": 0.18615815767028146, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0002144142857142857, | |
| "loss": 0.2697, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_loss": 0.19151142239570618, | |
| "eval_runtime": 183.9907, | |
| "eval_samples_per_second": 21.735, | |
| "eval_steps_per_second": 2.718, | |
| "eval_wer": 0.18472036066772268, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.00019298571428571424, | |
| "loss": 0.2363, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "eval_loss": 0.166675865650177, | |
| "eval_runtime": 184.6, | |
| "eval_samples_per_second": 21.663, | |
| "eval_steps_per_second": 2.709, | |
| "eval_wer": 0.16083830876081395, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.00017155714285714283, | |
| "loss": 0.1795, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 0.1457526981830597, | |
| "eval_runtime": 183.1764, | |
| "eval_samples_per_second": 21.831, | |
| "eval_steps_per_second": 2.73, | |
| "eval_wer": 0.14287803094918972, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.00015012857142857142, | |
| "loss": 0.1636, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "eval_loss": 0.14684966206550598, | |
| "eval_runtime": 182.9497, | |
| "eval_samples_per_second": 21.858, | |
| "eval_steps_per_second": 2.733, | |
| "eval_wer": 0.1388083343487267, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "eval_loss": 0.13510963320732117, | |
| "eval_runtime": 182.9435, | |
| "eval_samples_per_second": 21.859, | |
| "eval_steps_per_second": 2.733, | |
| "eval_wer": 0.1314487632508834, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.00012869999999999998, | |
| "loss": 0.1445, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 0.11629413068294525, | |
| "eval_runtime": 183.428, | |
| "eval_samples_per_second": 21.801, | |
| "eval_steps_per_second": 2.726, | |
| "eval_wer": 0.11078347751919093, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.00010727142857142857, | |
| "loss": 0.1153, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "eval_loss": 0.10931886732578278, | |
| "eval_runtime": 183.2606, | |
| "eval_samples_per_second": 21.821, | |
| "eval_steps_per_second": 2.728, | |
| "eval_wer": 0.10876081393932009, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 8.584285714285713e-05, | |
| "loss": 0.1011, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "eval_loss": 0.12333084642887115, | |
| "eval_runtime": 183.8962, | |
| "eval_samples_per_second": 21.746, | |
| "eval_steps_per_second": 2.719, | |
| "eval_wer": 0.11409772145729255, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.441428571428571e-05, | |
| "loss": 0.0978, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.1147131398320198, | |
| "eval_runtime": 183.2548, | |
| "eval_samples_per_second": 21.822, | |
| "eval_steps_per_second": 2.728, | |
| "eval_wer": 0.10408188132082369, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "eval_loss": 0.11119665205478668, | |
| "eval_runtime": 183.6945, | |
| "eval_samples_per_second": 21.77, | |
| "eval_steps_per_second": 2.722, | |
| "eval_wer": 0.09840380163275253, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 4.2985714285714285e-05, | |
| "loss": 0.0748, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "eval_loss": 0.11000872403383255, | |
| "eval_runtime": 184.0303, | |
| "eval_samples_per_second": 21.73, | |
| "eval_steps_per_second": 2.717, | |
| "eval_wer": 0.09382234677714146, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 2.1557142857142854e-05, | |
| "loss": 0.075, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_loss": 0.1180349737405777, | |
| "eval_runtime": 183.7553, | |
| "eval_samples_per_second": 21.763, | |
| "eval_steps_per_second": 2.721, | |
| "eval_wer": 0.09601559644206166, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 1.7142857142857143e-07, | |
| "loss": 0.0668, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 7500, | |
| "total_flos": 7.948903600818893e+18, | |
| "train_loss": 0.4369275619506836, | |
| "train_runtime": 9373.6438, | |
| "train_samples_per_second": 6.4, | |
| "train_steps_per_second": 0.8 | |
| } | |
| ], | |
| "max_steps": 7500, | |
| "num_train_epochs": 10, | |
| "total_flos": 7.948903600818893e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |