| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.8, | |
| "global_step": 5200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002394, | |
| "loss": 3.4203, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1205.6526, | |
| "eval_samples_per_second": 3.822, | |
| "eval_steps_per_second": 0.478, | |
| "eval_wer": 0.5028872209091932, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0002873142857142857, | |
| "loss": 0.4156, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1205.8762, | |
| "eval_samples_per_second": 3.821, | |
| "eval_steps_per_second": 0.478, | |
| "eval_wer": 0.21501354872780476, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0002702142857142857, | |
| "loss": 0.248, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1206.2112, | |
| "eval_samples_per_second": 3.82, | |
| "eval_steps_per_second": 0.478, | |
| "eval_wer": 0.19620116181423092, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00025315714285714284, | |
| "loss": 0.1884, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1198.1731, | |
| "eval_samples_per_second": 3.846, | |
| "eval_steps_per_second": 0.481, | |
| "eval_wer": 0.18049675782839433, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.00023605714285714283, | |
| "loss": 0.1488, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1195.3799, | |
| "eval_samples_per_second": 3.855, | |
| "eval_steps_per_second": 0.482, | |
| "eval_wer": 0.18608073829744867, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.00021899999999999998, | |
| "loss": 0.1232, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1196.1097, | |
| "eval_samples_per_second": 3.852, | |
| "eval_steps_per_second": 0.482, | |
| "eval_wer": 0.17618540546624073, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 0.0002019, | |
| "loss": 0.107, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1192.8335, | |
| "eval_samples_per_second": 3.863, | |
| "eval_steps_per_second": 0.483, | |
| "eval_wer": 0.17763118025435248, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.0001848428571428571, | |
| "loss": 0.2435, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1189.2109, | |
| "eval_samples_per_second": 3.875, | |
| "eval_steps_per_second": 0.484, | |
| "eval_wer": 0.5853483278359262, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 0.00016774285714285715, | |
| "loss": 0.7331, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1196.7777, | |
| "eval_samples_per_second": 3.85, | |
| "eval_steps_per_second": 0.481, | |
| "eval_wer": 0.5020128301690777, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.0001506857142857143, | |
| "loss": 0.5402, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1194.9864, | |
| "eval_samples_per_second": 3.856, | |
| "eval_steps_per_second": 0.482, | |
| "eval_wer": 0.29286895393432544, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 0.00013358571428571426, | |
| "loss": 1.3361, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1207.0654, | |
| "eval_samples_per_second": 3.818, | |
| "eval_steps_per_second": 0.477, | |
| "eval_wer": 0.9734566137703555, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 0.0001165285714285714, | |
| "loss": 0.9035, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1210.7223, | |
| "eval_samples_per_second": 3.806, | |
| "eval_steps_per_second": 0.476, | |
| "eval_wer": 0.4126085413257841, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 9.947142857142856e-05, | |
| "loss": 0.5824, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 1209.2109, | |
| "eval_samples_per_second": 3.811, | |
| "eval_steps_per_second": 0.476, | |
| "eval_wer": 0.2804110502211949, | |
| "step": 5200 | |
| } | |
| ], | |
| "max_steps": 7500, | |
| "num_train_epochs": 30, | |
| "total_flos": 8.725540278590487e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |