| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 200.0, | |
| "global_step": 51600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 9.475479396164832e-05, | |
| "loss": 27.7884, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.9939393939393939, | |
| "eval_loss": 3.886181116104126, | |
| "eval_runtime": 7.6119, | |
| "eval_samples_per_second": 27.588, | |
| "eval_steps_per_second": 3.547, | |
| "eval_wer": 1.0, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 8.422847817217463e-05, | |
| "loss": 2.6676, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_cer": 0.9238095238095239, | |
| "eval_loss": 3.842160224914551, | |
| "eval_runtime": 7.6611, | |
| "eval_samples_per_second": 27.411, | |
| "eval_steps_per_second": 3.524, | |
| "eval_wer": 1.0, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 7.370379436964505e-05, | |
| "loss": 2.3606, | |
| "step": 19350 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_cer": 0.8813852813852814, | |
| "eval_loss": 3.7868971824645996, | |
| "eval_runtime": 7.5871, | |
| "eval_samples_per_second": 27.678, | |
| "eval_steps_per_second": 3.559, | |
| "eval_wer": 1.0, | |
| "step": 19350 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 6.318074255405957e-05, | |
| "loss": 2.2401, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_cer": 0.8701298701298701, | |
| "eval_loss": 3.7943010330200195, | |
| "eval_runtime": 7.6629, | |
| "eval_samples_per_second": 27.405, | |
| "eval_steps_per_second": 3.523, | |
| "eval_wer": 1.0, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "learning_rate": 5.2657690738474096e-05, | |
| "loss": 2.1734, | |
| "step": 32250 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "eval_cer": 0.8528138528138528, | |
| "eval_loss": 3.783539056777954, | |
| "eval_runtime": 7.5808, | |
| "eval_samples_per_second": 27.702, | |
| "eval_steps_per_second": 3.562, | |
| "eval_wer": 1.0, | |
| "step": 32250 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 4.213790289677683e-05, | |
| "loss": 2.1344, | |
| "step": 38700 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "eval_cer": 0.8502164502164502, | |
| "eval_loss": 3.778026819229126, | |
| "eval_runtime": 7.5814, | |
| "eval_samples_per_second": 27.699, | |
| "eval_steps_per_second": 3.561, | |
| "eval_wer": 1.0, | |
| "step": 38700 | |
| }, | |
| { | |
| "epoch": 175.0, | |
| "learning_rate": 3.1614851081191354e-05, | |
| "loss": 2.1088, | |
| "step": 45150 | |
| }, | |
| { | |
| "epoch": 175.0, | |
| "eval_cer": 0.8424242424242424, | |
| "eval_loss": 3.7703933715820312, | |
| "eval_runtime": 7.591, | |
| "eval_samples_per_second": 27.664, | |
| "eval_steps_per_second": 3.557, | |
| "eval_wer": 1.0, | |
| "step": 45150 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 2.109343125254998e-05, | |
| "loss": 2.0884, | |
| "step": 51600 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "eval_cer": 0.8398268398268398, | |
| "eval_loss": 3.760558605194092, | |
| "eval_runtime": 7.5804, | |
| "eval_samples_per_second": 27.703, | |
| "eval_steps_per_second": 3.562, | |
| "eval_wer": 1.0, | |
| "step": 51600 | |
| } | |
| ], | |
| "max_steps": 64500, | |
| "num_train_epochs": 250, | |
| "total_flos": 5.1534348266634125e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |