| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 2200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 1.1212499999999998e-05, | |
| "loss": 9.6445, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "eval_loss": 4.396285057067871, | |
| "eval_runtime": 12.5672, | |
| "eval_samples_per_second": 23.951, | |
| "eval_steps_per_second": 3.024, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 2.2462499999999997e-05, | |
| "loss": 3.6459, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "eval_loss": 3.2267072200775146, | |
| "eval_runtime": 12.4767, | |
| "eval_samples_per_second": 24.125, | |
| "eval_steps_per_second": 3.046, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "learning_rate": 3.37125e-05, | |
| "loss": 3.0978, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "eval_loss": 3.0927422046661377, | |
| "eval_runtime": 12.5507, | |
| "eval_samples_per_second": 23.983, | |
| "eval_steps_per_second": 3.028, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 54.55, | |
| "learning_rate": 4.4962499999999995e-05, | |
| "loss": 2.8357, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 54.55, | |
| "eval_loss": 2.146217107772827, | |
| "eval_runtime": 12.4871, | |
| "eval_samples_per_second": 24.105, | |
| "eval_steps_per_second": 3.043, | |
| "eval_wer": 1.002932551319648, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "learning_rate": 5.62125e-05, | |
| "loss": 1.2723, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "eval_loss": 0.6747255325317383, | |
| "eval_runtime": 12.3256, | |
| "eval_samples_per_second": 24.421, | |
| "eval_steps_per_second": 3.083, | |
| "eval_wer": 0.6996229576874738, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 81.82, | |
| "learning_rate": 6.746249999999999e-05, | |
| "loss": 0.6528, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 81.82, | |
| "eval_loss": 0.5928319096565247, | |
| "eval_runtime": 12.4668, | |
| "eval_samples_per_second": 24.144, | |
| "eval_steps_per_second": 3.048, | |
| "eval_wer": 0.6422287390029325, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 95.45, | |
| "learning_rate": 3.7875e-05, | |
| "loss": 0.4905, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 95.45, | |
| "eval_loss": 0.5586517453193665, | |
| "eval_runtime": 12.2287, | |
| "eval_samples_per_second": 24.614, | |
| "eval_steps_per_second": 3.107, | |
| "eval_wer": 0.5680770842061165, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 2200, | |
| "total_flos": 1.3980125790314312e+19, | |
| "train_loss": 2.969521954276345, | |
| "train_runtime": 5494.3742, | |
| "train_samples_per_second": 12.813, | |
| "train_steps_per_second": 0.4 | |
| } | |
| ], | |
| "max_steps": 2200, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.3980125790314312e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |