| { | |
| "best_metric": 23.97033275424295, | |
| "best_model_checkpoint": "whisper-small-english/checkpoint-4000", | |
| "epoch": 7.518796992481203, | |
| "eval_steps": 1000, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9398496240601504, | |
| "grad_norm": 3.8690667152404785, | |
| "learning_rate": 1.7465e-05, | |
| "loss": 0.5642, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8796992481203008, | |
| "grad_norm": 3.2477657794952393, | |
| "learning_rate": 1.555944444444444e-05, | |
| "loss": 0.3539, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.8796992481203008, | |
| "eval_loss": 0.4700809717178345, | |
| "eval_runtime": 321.0656, | |
| "eval_samples_per_second": 8.771, | |
| "eval_steps_per_second": 0.368, | |
| "eval_wer": 31.456635039422693, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.819548872180451, | |
| "grad_norm": 2.6958045959472656, | |
| "learning_rate": 1.3615e-05, | |
| "loss": 0.2419, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "grad_norm": 2.6407883167266846, | |
| "learning_rate": 1.1670555555555554e-05, | |
| "loss": 0.1559, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "eval_loss": 0.5331035256385803, | |
| "eval_runtime": 311.0119, | |
| "eval_samples_per_second": 9.054, | |
| "eval_steps_per_second": 0.379, | |
| "eval_wer": 26.08846719230255, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.6992481203007515, | |
| "grad_norm": 1.7738432884216309, | |
| "learning_rate": 9.72611111111111e-06, | |
| "loss": 0.0946, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.639097744360902, | |
| "grad_norm": 1.7158890962600708, | |
| "learning_rate": 7.781666666666665e-06, | |
| "loss": 0.0549, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.639097744360902, | |
| "eval_loss": 0.6581208109855652, | |
| "eval_runtime": 306.2753, | |
| "eval_samples_per_second": 9.194, | |
| "eval_steps_per_second": 0.385, | |
| "eval_wer": 24.360550581317653, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.578947368421053, | |
| "grad_norm": 0.39378559589385986, | |
| "learning_rate": 5.837222222222222e-06, | |
| "loss": 0.0303, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.518796992481203, | |
| "grad_norm": 0.6418510675430298, | |
| "learning_rate": 3.892777777777777e-06, | |
| "loss": 0.0163, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.518796992481203, | |
| "eval_loss": 0.758054256439209, | |
| "eval_runtime": 304.6518, | |
| "eval_samples_per_second": 9.243, | |
| "eval_steps_per_second": 0.387, | |
| "eval_wer": 23.97033275424295, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.535991451418624e+19, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |