| { |
| "best_metric": 0.27609055770292656, |
| "best_model_checkpoint": "./net-finetune/checkpoint-150", |
| "epoch": 2.003338898163606, |
| "eval_steps": 25, |
| "global_step": 150, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.33, |
| "learning_rate": 4.2000000000000004e-06, |
| "loss": 4.1307, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.33, |
| "eval_loss": 2.314058303833008, |
| "eval_runtime": 350.0185, |
| "eval_samples_per_second": 0.863, |
| "eval_steps_per_second": 0.431, |
| "eval_wer": 16.675869685256764, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.67, |
| "learning_rate": 9e-06, |
| "loss": 1.3074, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.67, |
| "eval_loss": 0.9148856401443481, |
| "eval_runtime": 344.4232, |
| "eval_samples_per_second": 0.877, |
| "eval_steps_per_second": 0.438, |
| "eval_wer": 0.5521811154058531, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 8.666666666666668e-06, |
| "loss": 0.7543, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.5279556512832642, |
| "eval_runtime": 364.0248, |
| "eval_samples_per_second": 0.83, |
| "eval_steps_per_second": 0.415, |
| "eval_wer": 0.6626173384870238, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.34, |
| "learning_rate": 7e-06, |
| "loss": 0.3644, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.34, |
| "eval_loss": 0.21674829721450806, |
| "eval_runtime": 344.0393, |
| "eval_samples_per_second": 0.878, |
| "eval_steps_per_second": 0.439, |
| "eval_wer": 2.5400331308669246, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.67, |
| "learning_rate": 5.333333333333334e-06, |
| "loss": 0.1453, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.67, |
| "eval_loss": 0.0939120501279831, |
| "eval_runtime": 346.6619, |
| "eval_samples_per_second": 0.871, |
| "eval_steps_per_second": 0.436, |
| "eval_wer": 1.8221976808393152, |
| "step": 125 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 3.6666666666666666e-06, |
| "loss": 0.0758, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.049126043915748596, |
| "eval_runtime": 345.3745, |
| "eval_samples_per_second": 0.874, |
| "eval_steps_per_second": 0.437, |
| "eval_wer": 0.27609055770292656, |
| "step": 150 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 200, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 50, |
| "total_flos": 2.449451778048e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
