{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 189, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.5873015873015874, "grad_norm": 2.6376938440744495, "learning_rate": 1.092268359463302e-06, "loss": 1.1317, "step": 100 }, { "epoch": 3.0, "step": 189, "total_flos": 268701811081216.0, "train_loss": 1.001537181712963, "train_runtime": 3023.1737, "train_samples_per_second": 15.95, "train_steps_per_second": 0.063 } ], "logging_steps": 100, "max_steps": 189, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 268701811081216.0, "train_batch_size": 32, "trial_name": null, "trial_params": null }