{ "best_metric": 0.4638437032699585, "best_model_checkpoint": "/content/Development/whisperfinetune/whisper-medium-20241128102840/20241128102840/checkpoint-16", "epoch": 0.29836829836829837, "eval_steps": 1, "global_step": 16, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.018648018648018648, "grad_norm": 3.2345731258392334, "learning_rate": 1e-05, "loss": 0.6187, "step": 1 }, { "epoch": 0.018648018648018648, "eval_loss": 0.6093018651008606, "eval_model_preparation_time": 0.0128, "eval_runtime": 6393.4078, "eval_samples_per_second": 1.431, "eval_steps_per_second": 0.179, "eval_wer": 6.035647060074655, "step": 1 }, { "epoch": 0.037296037296037296, "grad_norm": 3.4020540714263916, "learning_rate": 1e-05, "loss": 0.6045, "step": 2 }, { "epoch": 0.037296037296037296, "eval_loss": 0.6032565832138062, "eval_model_preparation_time": 0.0128, "eval_runtime": 6357.2352, "eval_samples_per_second": 1.439, "eval_steps_per_second": 0.18, "eval_wer": 6.098390990396988, "step": 2 }, { "epoch": 0.055944055944055944, "grad_norm": 3.5230391025543213, "learning_rate": 1e-05, "loss": 0.6131, "step": 3 }, { "epoch": 0.055944055944055944, "eval_loss": 0.588996410369873, "eval_model_preparation_time": 0.0128, "eval_runtime": 6406.215, "eval_samples_per_second": 1.428, "eval_steps_per_second": 0.179, "eval_wer": 5.845288356215371, "step": 3 }, { "epoch": 0.07459207459207459, "grad_norm": 3.259377956390381, "learning_rate": 1e-05, "loss": 0.5727, "step": 4 }, { "epoch": 0.07459207459207459, "eval_loss": 0.5783398747444153, "eval_model_preparation_time": 0.0128, "eval_runtime": 6393.9371, "eval_samples_per_second": 1.431, "eval_steps_per_second": 0.179, "eval_wer": 5.747982091393447, "step": 4 }, { "epoch": 0.09324009324009325, "grad_norm": 3.2146058082580566, "learning_rate": 1e-05, "loss": 0.5811, "step": 5 }, { "epoch": 0.09324009324009325, "eval_loss": 0.5736284852027893, "eval_model_preparation_time": 0.0128, "eval_runtime": 6384.5096, "eval_samples_per_second": 1.433, "eval_steps_per_second": 0.179, "eval_wer": 5.971307945083109, "step": 5 }, { "epoch": 0.11188811188811189, "grad_norm": 3.293224573135376, "learning_rate": 1e-05, "loss": 0.566, "step": 6 }, { "epoch": 0.11188811188811189, "eval_loss": 0.5640575289726257, "eval_model_preparation_time": 0.0128, "eval_runtime": 6374.3313, "eval_samples_per_second": 1.435, "eval_steps_per_second": 0.179, "eval_wer": 5.889421798730233, "step": 6 }, { "epoch": 0.13053613053613053, "grad_norm": 3.3218321800231934, "learning_rate": 1e-05, "loss": 0.5664, "step": 7 }, { "epoch": 0.13053613053613053, "eval_loss": 0.5516149997711182, "eval_model_preparation_time": 0.0128, "eval_runtime": 6395.0724, "eval_samples_per_second": 1.431, "eval_steps_per_second": 0.179, "eval_wer": 5.682579519955761, "step": 7 }, { "epoch": 0.14918414918414918, "grad_norm": 3.191657781600952, "learning_rate": 1e-05, "loss": 0.5603, "step": 8 }, { "epoch": 0.14918414918414918, "eval_loss": 0.5415236353874207, "eval_model_preparation_time": 0.0128, "eval_runtime": 6394.4097, "eval_samples_per_second": 1.431, "eval_steps_per_second": 0.179, "eval_wer": 5.577829060010847, "step": 8 }, { "epoch": 0.16783216783216784, "grad_norm": 3.148787260055542, "learning_rate": 1e-05, "loss": 0.5269, "step": 9 }, { "epoch": 0.16783216783216784, "eval_loss": 0.5326558351516724, "eval_model_preparation_time": 0.0128, "eval_runtime": 6400.9663, "eval_samples_per_second": 1.429, "eval_steps_per_second": 0.179, "eval_wer": 5.527314878819138, "step": 9 }, { "epoch": 0.1864801864801865, "grad_norm": 3.1820664405822754, "learning_rate": 1e-05, "loss": 0.5259, "step": 10 }, { "epoch": 0.1864801864801865, "eval_loss": 0.5232037901878357, "eval_model_preparation_time": 0.0128, "eval_runtime": 6385.319, "eval_samples_per_second": 1.433, "eval_steps_per_second": 0.179, "eval_wer": 5.5602820286495165, "step": 10 }, { "epoch": 0.20512820512820512, "grad_norm": 3.181210994720459, "learning_rate": 1e-05, "loss": 0.5179, "step": 11 }, { "epoch": 0.20512820512820512, "eval_loss": 0.5133035778999329, "eval_model_preparation_time": 0.0128, "eval_runtime": 6399.1957, "eval_samples_per_second": 1.43, "eval_steps_per_second": 0.179, "eval_wer": 5.512958216796231, "step": 11 }, { "epoch": 0.22377622377622378, "grad_norm": 3.187267303466797, "learning_rate": 1e-05, "loss": 0.5182, "step": 12 }, { "epoch": 0.22377622377622378, "eval_loss": 0.5032967925071716, "eval_model_preparation_time": 0.0128, "eval_runtime": 6424.8662, "eval_samples_per_second": 1.424, "eval_steps_per_second": 0.178, "eval_wer": 5.4257547882126484, "step": 12 }, { "epoch": 0.24242424242424243, "grad_norm": 3.1449100971221924, "learning_rate": 1e-05, "loss": 0.5046, "step": 13 }, { "epoch": 0.24242424242424243, "eval_loss": 0.4931994378566742, "eval_model_preparation_time": 0.0128, "eval_runtime": 6390.9771, "eval_samples_per_second": 1.432, "eval_steps_per_second": 0.179, "eval_wer": 5.395446279497623, "step": 13 }, { "epoch": 0.26107226107226106, "grad_norm": 3.126314163208008, "learning_rate": 1e-05, "loss": 0.4841, "step": 14 }, { "epoch": 0.26107226107226106, "eval_loss": 0.4834253191947937, "eval_model_preparation_time": 0.0128, "eval_runtime": 6368.6339, "eval_samples_per_second": 1.437, "eval_steps_per_second": 0.18, "eval_wer": 5.402890474620612, "step": 14 }, { "epoch": 0.27972027972027974, "grad_norm": 3.1767537593841553, "learning_rate": 1e-05, "loss": 0.4895, "step": 15 }, { "epoch": 0.27972027972027974, "eval_loss": 0.4739968478679657, "eval_model_preparation_time": 0.0128, "eval_runtime": 6385.9501, "eval_samples_per_second": 1.433, "eval_steps_per_second": 0.179, "eval_wer": 5.3539714780981145, "step": 15 }, { "epoch": 0.29836829836829837, "grad_norm": 3.1699957847595215, "learning_rate": 1e-05, "loss": 0.477, "step": 16 }, { "epoch": 0.29836829836829837, "eval_loss": 0.4638437032699585, "eval_model_preparation_time": 0.0128, "eval_runtime": 6377.959, "eval_samples_per_second": 1.435, "eval_steps_per_second": 0.179, "eval_wer": 5.3709867812363745, "step": 16 } ], "logging_steps": 1, "max_steps": 30, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 2, "stateful_callbacks": { "EarlyStoppingCallback": { "args": { "early_stopping_patience": 3, "early_stopping_threshold": 0.0 }, "attributes": { "early_stopping_patience_counter": 0 } }, "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 8.36079540240384e+18, "train_batch_size": 16, "trial_name": null, "trial_params": null }