{ "best_metric": null, "best_model_checkpoint": null, "epoch": 5.0, "global_step": 470, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 8e-05, "loss": 3.362, "step": 94 }, { "epoch": 1.0, "eval_loss": 2.299670934677124, "eval_runtime": 38.536, "eval_samples_per_second": 5.19, "eval_steps_per_second": 0.649, "step": 94 }, { "epoch": 2.0, "learning_rate": 6e-05, "loss": 3.0516, "step": 188 }, { "epoch": 2.0, "eval_loss": 2.316689968109131, "eval_runtime": 38.2069, "eval_samples_per_second": 5.235, "eval_steps_per_second": 0.654, "step": 188 }, { "epoch": 3.0, "learning_rate": 4e-05, "loss": 2.8505, "step": 282 }, { "epoch": 3.0, "eval_loss": 2.0240907669067383, "eval_runtime": 38.0759, "eval_samples_per_second": 5.253, "eval_steps_per_second": 0.657, "step": 282 }, { "epoch": 4.0, "learning_rate": 2e-05, "loss": 2.5805, "step": 376 }, { "epoch": 4.0, "eval_loss": 1.9035601615905762, "eval_runtime": 38.7284, "eval_samples_per_second": 5.164, "eval_steps_per_second": 0.646, "step": 376 }, { "epoch": 5.0, "learning_rate": 0.0, "loss": 2.5158, "step": 470 }, { "epoch": 5.0, "eval_loss": 1.8688610792160034, "eval_runtime": 38.739, "eval_samples_per_second": 5.163, "eval_steps_per_second": 0.645, "step": 470 }, { "epoch": 5.0, "step": 470, "total_flos": 2.6103004704e+18, "train_loss": 2.8720729097406914, "train_runtime": 2070.4092, "train_samples_per_second": 1.811, "train_steps_per_second": 0.227 } ], "max_steps": 470, "num_train_epochs": 5, "total_flos": 2.6103004704e+18, "trial_name": null, "trial_params": null }