{ "best_metric": 72.41379310344827, "best_model_checkpoint": "../openai/whisper-large-v3-finetuned/checkpoint-20", "epoch": 7.09, "eval_steps": 20, "global_step": 100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.07, "grad_norm": 1.2736600637435913, "learning_rate": 0.0001, "loss": 0.054, "step": 20 }, { "epoch": 1.07, "eval_loss": 2.1966075897216797, "eval_runtime": 68.9209, "eval_samples_per_second": 0.145, "eval_steps_per_second": 0.044, "eval_wer": 72.41379310344827, "step": 20 }, { "epoch": 3.01, "grad_norm": 0.4058435261249542, "learning_rate": 0.0001, "loss": 0.0226, "step": 40 }, { "epoch": 3.01, "eval_loss": 2.285231351852417, "eval_runtime": 69.6442, "eval_samples_per_second": 0.144, "eval_steps_per_second": 0.043, "eval_wer": 79.3103448275862, "step": 40 }, { "epoch": 4.08, "grad_norm": 0.26837655901908875, "learning_rate": 0.0001, "loss": 0.0124, "step": 60 }, { "epoch": 4.08, "eval_loss": 2.316169023513794, "eval_runtime": 217.2614, "eval_samples_per_second": 0.046, "eval_steps_per_second": 0.014, "eval_wer": 147.1264367816092, "step": 60 }, { "epoch": 6.02, "grad_norm": 0.08057563006877899, "learning_rate": 0.0001, "loss": 0.0061, "step": 80 }, { "epoch": 6.02, "eval_loss": 2.2900302410125732, "eval_runtime": 68.7884, "eval_samples_per_second": 0.145, "eval_steps_per_second": 0.044, "eval_wer": 74.71264367816092, "step": 80 }, { "epoch": 7.09, "grad_norm": 0.15883919596672058, "learning_rate": 0.0001, "loss": 0.0039, "step": 100 }, { "epoch": 7.09, "eval_loss": 2.3415772914886475, "eval_runtime": 216.593, "eval_samples_per_second": 0.046, "eval_steps_per_second": 0.014, "eval_wer": 141.3793103448276, "step": 100 } ], "logging_steps": 20, "max_steps": 100, "num_input_tokens_seen": 0, "num_train_epochs": 9223372036854775807, "save_steps": 20, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.32542251794432e+18, "train_batch_size": 4, "trial_name": null, "trial_params": null }