{ "best_metric": 0.4769304327038011, "best_model_checkpoint": "w2v-bert-2.0-CV_Fleurs-lg-5hrs-v4/checkpoint-1290", "epoch": 6.0, "eval_steps": 500, "global_step": 1548, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 11.246596336364746, "learning_rate": 5.1200000000000004e-05, "loss": 1.9959, "step": 258 }, { "epoch": 1.0, "eval_cer": 0.13379445701894616, "eval_loss": 0.5475470423698425, "eval_runtime": 29.6601, "eval_samples_per_second": 27.849, "eval_steps_per_second": 6.979, "eval_wer": 0.6078173559646187, "step": 258 }, { "epoch": 2.0, "grad_norm": 1.1338785886764526, "learning_rate": 9.994466403162056e-05, "loss": 0.5447, "step": 516 }, { "epoch": 2.0, "eval_cer": 0.1309612371199291, "eval_loss": 0.5649892687797546, "eval_runtime": 29.1014, "eval_samples_per_second": 28.384, "eval_steps_per_second": 7.113, "eval_wer": 0.5978962467128854, "step": 516 }, { "epoch": 3.0, "grad_norm": 5.585188388824463, "learning_rate": 9.892490118577076e-05, "loss": 0.4574, "step": 774 }, { "epoch": 3.0, "eval_cer": 0.11218917678342487, "eval_loss": 0.45441317558288574, "eval_runtime": 29.5398, "eval_samples_per_second": 27.962, "eval_steps_per_second": 7.007, "eval_wer": 0.5168539325842697, "step": 774 }, { "epoch": 4.0, "grad_norm": 3.5034706592559814, "learning_rate": 9.790513833992096e-05, "loss": 0.3712, "step": 1032 }, { "epoch": 4.0, "eval_cer": 0.10859621076623562, "eval_loss": 0.4465671181678772, "eval_runtime": 29.2068, "eval_samples_per_second": 28.281, "eval_steps_per_second": 7.087, "eval_wer": 0.5049007889074827, "step": 1032 }, { "epoch": 5.0, "grad_norm": 17.74321174621582, "learning_rate": 9.688932806324111e-05, "loss": 0.3187, "step": 1290 }, { "epoch": 5.0, "eval_cer": 0.10036562781937036, "eval_loss": 0.44379866123199463, "eval_runtime": 29.231, "eval_samples_per_second": 28.258, "eval_steps_per_second": 7.082, "eval_wer": 0.4769304327038011, "step": 1290 }, { "epoch": 6.0, "grad_norm": 10.85033893585205, "learning_rate": 9.586956521739131e-05, "loss": 0.276, "step": 1548 }, { "epoch": 6.0, "eval_cer": 0.1056996786907042, "eval_loss": 0.4188626706600189, "eval_runtime": 29.2483, "eval_samples_per_second": 28.241, "eval_steps_per_second": 7.077, "eval_wer": 0.4794405928759264, "step": 1548 } ], "logging_steps": 500, "max_steps": 25800, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { "args": { "early_stopping_patience": 20, "early_stopping_threshold": 0.001 }, "attributes": { "early_stopping_patience_counter": 1 } }, "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 3.2752161886451635e+18, "train_batch_size": 8, "trial_name": null, "trial_params": null }