| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 1580, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.00020544303797468352, | |
| "loss": 2.9089, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "eval_loss": 2.89827823638916, | |
| "eval_runtime": 29.5705, | |
| "eval_samples_per_second": 19.749, | |
| "eval_steps_per_second": 2.469, | |
| "eval_wer": 1.0002232641214557, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "learning_rate": 0.00011069620253164557, | |
| "loss": 2.5754, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "eval_loss": 1.8709616661071777, | |
| "eval_runtime": 29.8131, | |
| "eval_samples_per_second": 19.589, | |
| "eval_steps_per_second": 2.449, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 1.5759493670886074e-05, | |
| "loss": 1.4093, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_loss": 0.7194529175758362, | |
| "eval_runtime": 29.359, | |
| "eval_samples_per_second": 19.892, | |
| "eval_steps_per_second": 2.486, | |
| "eval_wer": 0.854655056932351, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1580, | |
| "total_flos": 3.6376414033360364e+18, | |
| "train_loss": 2.2397680982758725, | |
| "train_runtime": 1536.9705, | |
| "train_samples_per_second": 16.266, | |
| "train_steps_per_second": 1.028 | |
| } | |
| ], | |
| "max_steps": 1580, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.6376414033360364e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |