{ "best_metric": 1.1350098848342896, "best_model_checkpoint": "ckpt_base/checkpoint-1308", "epoch": 6.0, "global_step": 1308, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "eval_loss": 1.3215606212615967, "eval_runtime": 1149.9926, "eval_samples_per_second": 2.767, "eval_steps_per_second": 1.383, "eval_wer": 458.45417429255247, "step": 218 }, { "epoch": 2.0, "eval_loss": 1.2341229915618896, "eval_runtime": 804.821, "eval_samples_per_second": 3.954, "eval_steps_per_second": 1.977, "eval_wer": 131.4655779248205, "step": 436 }, { "epoch": 2.29, "learning_rate": 1.2450000000000001e-05, "loss": 1.7076, "step": 500 }, { "epoch": 3.0, "eval_loss": 1.1820706129074097, "eval_runtime": 647.2645, "eval_samples_per_second": 4.916, "eval_steps_per_second": 2.458, "eval_wer": 130.46600028157116, "step": 654 }, { "epoch": 4.0, "eval_loss": 1.1853020191192627, "eval_runtime": 1558.9928, "eval_samples_per_second": 2.041, "eval_steps_per_second": 1.021, "eval_wer": 117.52780515275236, "step": 872 }, { "epoch": 4.59, "learning_rate": 2.495e-05, "loss": 1.136, "step": 1000 }, { "epoch": 5.0, "eval_loss": 1.1510676145553589, "eval_runtime": 668.7602, "eval_samples_per_second": 4.758, "eval_steps_per_second": 2.379, "eval_wer": 130.6490215401943, "step": 1090 }, { "epoch": 6.0, "eval_loss": 1.1350098848342896, "eval_runtime": 656.7286, "eval_samples_per_second": 4.845, "eval_steps_per_second": 2.423, "eval_wer": 128.88920174574125, "step": 1308 } ], "max_steps": 4360, "num_train_epochs": 20, "total_flos": 2.704661904109863e+18, "trial_name": null, "trial_params": null }