{ "best_metric": 2.6429617404937744, "best_model_checkpoint": "miner_id_24/checkpoint-10", "epoch": 0.00024053784261608956, "eval_steps": 5, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 2.4053784261608958e-05, "eval_loss": 3.604696750640869, "eval_runtime": 2235.6462, "eval_samples_per_second": 15.66, "eval_steps_per_second": 3.915, "step": 1 }, { "epoch": 7.216135278482687e-05, "grad_norm": 21.76235580444336, "learning_rate": 0.00012, "loss": 10.3252, "step": 3 }, { "epoch": 0.00012026892130804478, "eval_loss": 3.373138666152954, "eval_runtime": 2235.6131, "eval_samples_per_second": 15.66, "eval_steps_per_second": 3.915, "step": 5 }, { "epoch": 0.00014432270556965373, "grad_norm": 12.278535842895508, "learning_rate": 0.0001992114701314478, "loss": 10.0076, "step": 6 }, { "epoch": 0.0002164840583544806, "grad_norm": 6.428678512573242, "learning_rate": 0.00018763066800438636, "loss": 9.4451, "step": 9 }, { "epoch": 0.00024053784261608956, "eval_loss": 2.6429617404937744, "eval_runtime": 2235.5851, "eval_samples_per_second": 15.66, "eval_steps_per_second": 3.915, "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": 6534646443540480.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }