| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 26.395939086294415, | |
| "global_step": 5200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2479, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 0.47397056221961975, | |
| "eval_runtime": 121.9231, | |
| "eval_samples_per_second": 17.47, | |
| "eval_steps_per_second": 2.19, | |
| "eval_wer": 0.5506920619898261, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 0.00018605221657613458, | |
| "loss": 0.2306, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "eval_loss": 0.4647212624549866, | |
| "eval_runtime": 123.4969, | |
| "eval_samples_per_second": 17.247, | |
| "eval_steps_per_second": 2.162, | |
| "eval_wer": 0.5497456524310895, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 0.00017206938847029964, | |
| "loss": 0.201, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "eval_loss": 0.43976637721061707, | |
| "eval_runtime": 125.3202, | |
| "eval_samples_per_second": 16.996, | |
| "eval_steps_per_second": 2.131, | |
| "eval_wer": 0.5441854962735123, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0001580515156824952, | |
| "loss": 0.2676, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "eval_loss": 0.46590062975883484, | |
| "eval_runtime": 127.6886, | |
| "eval_samples_per_second": 16.681, | |
| "eval_steps_per_second": 2.091, | |
| "eval_wer": 0.5331834851532, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "learning_rate": 0.00014403364289469075, | |
| "loss": 0.2295, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "eval_loss": 0.433741956949234, | |
| "eval_runtime": 126.2574, | |
| "eval_samples_per_second": 16.87, | |
| "eval_steps_per_second": 2.115, | |
| "eval_wer": 0.5302259552821483, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.18, | |
| "learning_rate": 0.00013001577010688628, | |
| "loss": 0.2011, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 12.18, | |
| "eval_loss": 0.43763187527656555, | |
| "eval_runtime": 127.2021, | |
| "eval_samples_per_second": 16.745, | |
| "eval_steps_per_second": 2.099, | |
| "eval_wer": 0.5218265704483616, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 0.00011599789731908184, | |
| "loss": 0.1883, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "eval_loss": 0.4427320659160614, | |
| "eval_runtime": 133.553, | |
| "eval_samples_per_second": 15.949, | |
| "eval_steps_per_second": 1.999, | |
| "eval_wer": 0.5275050278007808, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "learning_rate": 0.00010198002453127737, | |
| "loss": 0.1599, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "eval_loss": 0.4528989791870117, | |
| "eval_runtime": 130.4751, | |
| "eval_samples_per_second": 16.325, | |
| "eval_steps_per_second": 2.046, | |
| "eval_wer": 0.5167396190701526, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 18.27, | |
| "learning_rate": 8.796215174347293e-05, | |
| "loss": 0.1473, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 18.27, | |
| "eval_loss": 0.46878582239151, | |
| "eval_runtime": 130.7961, | |
| "eval_samples_per_second": 16.285, | |
| "eval_steps_per_second": 2.041, | |
| "eval_wer": 0.5125990772506802, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 20.3, | |
| "learning_rate": 7.394427895566848e-05, | |
| "loss": 0.1306, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 20.3, | |
| "eval_loss": 0.46525096893310547, | |
| "eval_runtime": 130.9702, | |
| "eval_samples_per_second": 16.263, | |
| "eval_steps_per_second": 2.039, | |
| "eval_wer": 0.5118892700816279, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.34, | |
| "learning_rate": 5.9926406167864026e-05, | |
| "loss": 0.1222, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 22.34, | |
| "eval_loss": 0.4552956223487854, | |
| "eval_runtime": 132.9038, | |
| "eval_samples_per_second": 16.027, | |
| "eval_steps_per_second": 2.009, | |
| "eval_wer": 0.5014787649355259, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 24.37, | |
| "learning_rate": 4.5908533380059577e-05, | |
| "loss": 0.1097, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 24.37, | |
| "eval_loss": 0.4868864417076111, | |
| "eval_runtime": 128.1956, | |
| "eval_samples_per_second": 16.615, | |
| "eval_steps_per_second": 2.083, | |
| "eval_wer": 0.5012421625458418, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 26.4, | |
| "learning_rate": 3.189066059225513e-05, | |
| "loss": 0.0988, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 26.4, | |
| "eval_loss": 0.45241138339042664, | |
| "eval_runtime": 127.9868, | |
| "eval_samples_per_second": 16.642, | |
| "eval_steps_per_second": 2.086, | |
| "eval_wer": 0.4968650183366852, | |
| "step": 5200 | |
| } | |
| ], | |
| "max_steps": 6107, | |
| "num_train_epochs": 31, | |
| "total_flos": 1.2361810216998281e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |