| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0204081632653061, |
| "eval_steps": 30, |
| "global_step": 150, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.20408163265306123, |
| "grad_norm": 28829.30859375, |
| "learning_rate": 1e-05, |
| "loss": 3.6903, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.20408163265306123, |
| "eval_cer": 0.43646314994809854, |
| "eval_loss": 2.987450122833252, |
| "eval_runtime": 203.3511, |
| "eval_samples_per_second": 5.586, |
| "eval_steps_per_second": 0.349, |
| "eval_wer": 1.0127242651736863, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.40816326530612246, |
| "grad_norm": 26780.62109375, |
| "learning_rate": 7.500000000000001e-06, |
| "loss": 2.533, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.40816326530612246, |
| "eval_cer": 0.2920826649051618, |
| "eval_loss": 2.235991954803467, |
| "eval_runtime": 163.1323, |
| "eval_samples_per_second": 6.964, |
| "eval_steps_per_second": 0.435, |
| "eval_wer": 0.8878992238198244, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.6122448979591837, |
| "grad_norm": 9.42530632019043, |
| "learning_rate": 5e-06, |
| "loss": 2.0604, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.6122448979591837, |
| "eval_cer": 0.26696234783429273, |
| "eval_loss": 1.951379656791687, |
| "eval_runtime": 171.6219, |
| "eval_samples_per_second": 6.619, |
| "eval_steps_per_second": 0.414, |
| "eval_wer": 0.8252958391652883, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.8163265306122449, |
| "grad_norm": 724720.0625, |
| "learning_rate": 2.5e-06, |
| "loss": 1.852, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.8163265306122449, |
| "eval_cer": 0.25809191280551097, |
| "eval_loss": 1.8181612491607666, |
| "eval_runtime": 180.4927, |
| "eval_samples_per_second": 6.294, |
| "eval_steps_per_second": 0.393, |
| "eval_wer": 0.7948848454001781, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.0204081632653061, |
| "grad_norm": 9.10916519165039, |
| "learning_rate": 0.0, |
| "loss": 1.7929, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.0204081632653061, |
| "eval_cer": 0.253486835896952, |
| "eval_loss": 1.7784144878387451, |
| "eval_runtime": 183.691, |
| "eval_samples_per_second": 6.184, |
| "eval_steps_per_second": 0.387, |
| "eval_wer": 0.7835602493955974, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.0204081632653061, |
| "step": 150, |
| "total_flos": 2.3614434607104e+17, |
| "train_loss": 2.385703277587891, |
| "train_runtime": 5094.7107, |
| "train_samples_per_second": 1.884, |
| "train_steps_per_second": 0.029 |
| } |
| ], |
| "logging_steps": 30, |
| "max_steps": 150, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 30, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.3614434607104e+17, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
