{ "best_metric": 37.030138841855745, "best_model_checkpoint": "./whisper-tiny-hi/checkpoint-4500", "epoch": 9.0, "eval_steps": 500, "global_step": 4500, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 8.432821273803711, "learning_rate": 9.920000000000002e-06, "loss": 1.1322, "step": 500 }, { "epoch": 1.0, "eval_loss": 0.4738043546676636, "eval_runtime": 281.8939, "eval_samples_per_second": 1.774, "eval_steps_per_second": 1.774, "eval_wer": 56.50186251269895, "step": 500 }, { "epoch": 2.0, "grad_norm": 7.294736385345459, "learning_rate": 9.89979797979798e-06, "loss": 0.3933, "step": 1000 }, { "epoch": 2.0, "eval_loss": 0.34950900077819824, "eval_runtime": 279.7119, "eval_samples_per_second": 1.788, "eval_steps_per_second": 1.788, "eval_wer": 46.79986454453098, "step": 1000 }, { "epoch": 3.0, "grad_norm": 5.787722587585449, "learning_rate": 9.79878787878788e-06, "loss": 0.2912, "step": 1500 }, { "epoch": 3.0, "eval_loss": 0.3126949667930603, "eval_runtime": 279.6788, "eval_samples_per_second": 1.788, "eval_steps_per_second": 1.788, "eval_wer": 43.294954283779205, "step": 1500 }, { "epoch": 4.0, "grad_norm": 5.6364569664001465, "learning_rate": 9.697777777777778e-06, "loss": 0.233, "step": 2000 }, { "epoch": 4.0, "eval_loss": 0.28942596912384033, "eval_runtime": 275.6431, "eval_samples_per_second": 1.814, "eval_steps_per_second": 1.814, "eval_wer": 42.65154080596004, "step": 2000 }, { "epoch": 5.0, "grad_norm": 5.897181987762451, "learning_rate": 9.596767676767677e-06, "loss": 0.189, "step": 2500 }, { "epoch": 5.0, "eval_loss": 0.2822072207927704, "eval_runtime": 276.7489, "eval_samples_per_second": 1.807, "eval_steps_per_second": 1.807, "eval_wer": 42.38062986793092, "step": 2500 }, { "epoch": 6.0, "grad_norm": 6.849637508392334, "learning_rate": 9.495757575757577e-06, "loss": 0.1531, "step": 3000 }, { "epoch": 6.0, "eval_loss": 0.285165399312973, "eval_runtime": 279.3412, "eval_samples_per_second": 1.79, "eval_steps_per_second": 1.79, "eval_wer": 47.13850321706739, "step": 3000 }, { "epoch": 7.0, "grad_norm": 4.602267265319824, "learning_rate": 9.394747474747476e-06, "loss": 0.1224, "step": 3500 }, { "epoch": 7.0, "eval_loss": 0.29211270809173584, "eval_runtime": 280.3174, "eval_samples_per_second": 1.784, "eval_steps_per_second": 1.784, "eval_wer": 51.3376227565188, "step": 3500 }, { "epoch": 8.0, "grad_norm": 6.410825729370117, "learning_rate": 9.293737373737376e-06, "loss": 0.0964, "step": 4000 }, { "epoch": 8.0, "eval_loss": 0.30172762274742126, "eval_runtime": 278.8186, "eval_samples_per_second": 1.793, "eval_steps_per_second": 1.793, "eval_wer": 44.14155096512022, "step": 4000 }, { "epoch": 9.0, "grad_norm": 5.8979997634887695, "learning_rate": 9.192727272727273e-06, "loss": 0.0744, "step": 4500 }, { "epoch": 9.0, "eval_loss": 0.31876111030578613, "eval_runtime": 278.6007, "eval_samples_per_second": 1.795, "eval_steps_per_second": 1.795, "eval_wer": 37.030138841855745, "step": 4500 } ], "logging_steps": 500, "max_steps": 50000, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.77255972864e+18, "train_batch_size": 16, "trial_name": null, "trial_params": null }