{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "global_step": 11187, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.13, "learning_rate": 4.776526325198892e-05, "loss": 7.5403, "step": 500 }, { "epoch": 0.27, "learning_rate": 4.553052650397783e-05, "loss": 4.6339, "step": 1000 }, { "epoch": 0.4, "learning_rate": 4.329578975596675e-05, "loss": 3.9715, "step": 1500 }, { "epoch": 0.54, "learning_rate": 4.106105300795566e-05, "loss": 3.6273, "step": 2000 }, { "epoch": 0.67, "learning_rate": 3.882631625994458e-05, "loss": 3.44, "step": 2500 }, { "epoch": 0.8, "learning_rate": 3.659157951193349e-05, "loss": 3.2674, "step": 3000 }, { "epoch": 0.94, "learning_rate": 3.435684276392241e-05, "loss": 3.1397, "step": 3500 }, { "epoch": 1.07, "learning_rate": 3.212210601591133e-05, "loss": 3.0314, "step": 4000 }, { "epoch": 1.21, "learning_rate": 2.9887369267900244e-05, "loss": 2.9436, "step": 4500 }, { "epoch": 1.34, "learning_rate": 2.765263251988916e-05, "loss": 2.8667, "step": 5000 }, { "epoch": 1.47, "learning_rate": 2.5417895771878074e-05, "loss": 2.8367, "step": 5500 }, { "epoch": 1.61, "learning_rate": 2.318315902386699e-05, "loss": 2.7917, "step": 6000 }, { "epoch": 1.74, "learning_rate": 2.0948422275855904e-05, "loss": 2.7435, "step": 6500 }, { "epoch": 1.88, "learning_rate": 1.871368552784482e-05, "loss": 2.7113, "step": 7000 }, { "epoch": 2.01, "learning_rate": 1.6478948779833734e-05, "loss": 2.6678, "step": 7500 }, { "epoch": 2.15, "learning_rate": 1.4244212031822651e-05, "loss": 2.6282, "step": 8000 }, { "epoch": 2.28, "learning_rate": 1.2009475283811568e-05, "loss": 2.6228, "step": 8500 }, { "epoch": 2.41, "learning_rate": 9.774738535800483e-06, "loss": 2.6293, "step": 9000 }, { "epoch": 2.55, "learning_rate": 7.540001787789398e-06, "loss": 2.5696, "step": 9500 }, { "epoch": 2.68, "learning_rate": 5.305265039778315e-06, "loss": 2.5645, "step": 10000 }, { "epoch": 2.82, "learning_rate": 3.07052829176723e-06, "loss": 2.5531, "step": 10500 }, { "epoch": 2.95, "learning_rate": 8.357915437561455e-07, "loss": 2.537, "step": 11000 }, { "epoch": 3.0, "step": 11187, "total_flos": 2.440065492393984e+16, "train_loss": 3.1850600532033204, "train_runtime": 1975.4971, "train_samples_per_second": 33.976, "train_steps_per_second": 5.663 } ], "max_steps": 11187, "num_train_epochs": 3, "total_flos": 2.440065492393984e+16, "trial_name": null, "trial_params": null }