{ "best_metric": 172.4755700325733, "best_model_checkpoint": "./temp/checkpoint-70", "epoch": 6.571428571428571, "eval_steps": 10, "global_step": 100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.6349206349206349, "eval_cer": 186.56571539506893, "eval_loss": 9.826007843017578, "eval_runtime": 21.7924, "eval_samples_per_second": 2.294, "eval_steps_per_second": 0.184, "eval_wer": 332.2475570032573, "step": 10 }, { "epoch": 1.3174603174603174, "eval_cer": 186.54630168899243, "eval_loss": 8.942676544189453, "eval_runtime": 20.9166, "eval_samples_per_second": 2.39, "eval_steps_per_second": 0.191, "eval_wer": 332.5732899022801, "step": 20 }, { "epoch": 1.9523809523809523, "eval_cer": 179.30498932246167, "eval_loss": 7.41646671295166, "eval_runtime": 23.189, "eval_samples_per_second": 2.156, "eval_steps_per_second": 0.172, "eval_wer": 329.64169381107496, "step": 30 }, { "epoch": 2.571428571428571, "eval_cer": 173.6167734420501, "eval_loss": 4.776045322418213, "eval_runtime": 23.185, "eval_samples_per_second": 2.157, "eval_steps_per_second": 0.173, "eval_wer": 313.51791530944627, "step": 40 }, { "epoch": 3.1904761904761907, "grad_norm": 63047348.0, "learning_rate": 1.0000000000000002e-06, "loss": 6.244, "step": 50 }, { "epoch": 3.1904761904761907, "eval_cer": 173.05377596583187, "eval_loss": 1.3689286708831787, "eval_runtime": 20.2703, "eval_samples_per_second": 2.467, "eval_steps_per_second": 0.197, "eval_wer": 295.114006514658, "step": 50 }, { "epoch": 3.825396825396825, "eval_cer": 119.18074160357213, "eval_loss": 0.5696829557418823, "eval_runtime": 19.3264, "eval_samples_per_second": 2.587, "eval_steps_per_second": 0.207, "eval_wer": 184.85342019543972, "step": 60 }, { "epoch": 4.444444444444445, "eval_cer": 113.70607649000195, "eval_loss": 0.4699205756187439, "eval_runtime": 15.0547, "eval_samples_per_second": 3.321, "eval_steps_per_second": 0.266, "eval_wer": 172.4755700325733, "step": 70 }, { "epoch": 5.317460317460317, "eval_cer": 126.20850320326151, "eval_loss": 0.41475537419319153, "eval_runtime": 23.4851, "eval_samples_per_second": 2.129, "eval_steps_per_second": 0.17, "eval_wer": 175.2442996742671, "step": 80 }, { "epoch": 5.9523809523809526, "eval_cer": 127.43156668608037, "eval_loss": 0.37485015392303467, "eval_runtime": 19.6066, "eval_samples_per_second": 2.55, "eval_steps_per_second": 0.204, "eval_wer": 230.94462540716614, "step": 90 }, { "epoch": 6.571428571428571, "grad_norm": 471759.03125, "learning_rate": 2.0000000000000003e-06, "loss": 0.3668, "step": 100 }, { "epoch": 6.571428571428571, "eval_cer": 146.02989710735778, "eval_loss": 0.3358791768550873, "eval_runtime": 25.2622, "eval_samples_per_second": 1.979, "eval_steps_per_second": 0.158, "eval_wer": 242.01954397394138, "step": 100 } ], "logging_steps": 50, "max_steps": 105, "num_input_tokens_seen": 0, "num_train_epochs": 7, "save_steps": 50, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.5519745179648e+17, "train_batch_size": 16, "trial_name": null, "trial_params": null }