{ "best_metric": 1.3190627098083496, "best_model_checkpoint": "./outputs/checkpoint-200", "epoch": 0.26666666666666666, "eval_steps": 100, "global_step": 200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.13, "learning_rate": 0.0002, "loss": 1.2648, "step": 100 }, { "epoch": 0.13, "eval_loss": 1.3499459028244019, "eval_runtime": 466.0769, "eval_samples_per_second": 4.141, "eval_steps_per_second": 0.519, "step": 100 }, { "epoch": 0.27, "learning_rate": 0.0002, "loss": 1.1034, "step": 200 }, { "epoch": 0.27, "eval_loss": 1.3190627098083496, "eval_runtime": 466.6597, "eval_samples_per_second": 4.136, "eval_steps_per_second": 0.519, "step": 200 } ], "logging_steps": 100, "max_steps": 2250, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 100, "total_flos": 6647423649644544.0, "trial_name": null, "trial_params": null }