{ "best_metric": 11.5, "best_model_checkpoint": "miner_id_24/checkpoint-10", "epoch": 0.006351222610352493, "eval_steps": 5, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0006351222610352493, "eval_loss": 11.5, "eval_runtime": 5.8679, "eval_samples_per_second": 112.987, "eval_steps_per_second": 56.579, "step": 1 }, { "epoch": 0.001905366783105748, "grad_norm": 2.4608796593383886e-05, "learning_rate": 6e-05, "loss": 46.0, "step": 3 }, { "epoch": 0.0031756113051762463, "eval_loss": 11.5, "eval_runtime": 5.8543, "eval_samples_per_second": 113.25, "eval_steps_per_second": 56.711, "step": 5 }, { "epoch": 0.003810733566211496, "grad_norm": 2.0316416339483112e-05, "learning_rate": 9.96057350657239e-05, "loss": 46.0, "step": 6 }, { "epoch": 0.0057161003493172435, "grad_norm": 2.8792263037757948e-05, "learning_rate": 9.381533400219318e-05, "loss": 46.0, "step": 9 }, { "epoch": 0.006351222610352493, "eval_loss": 11.5, "eval_runtime": 5.8794, "eval_samples_per_second": 112.767, "eval_steps_per_second": 56.469, "step": 10 } ], "logging_steps": 3, "max_steps": 30, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 10, "stateful_callbacks": { "EarlyStoppingCallback": { "args": { "early_stopping_patience": 1, "early_stopping_threshold": 0.0 }, "attributes": { "early_stopping_patience_counter": 0 } }, "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 398960885760.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }