{ "best_metric": 11.931650161743164, "best_model_checkpoint": "miner_id_24/checkpoint-10", "epoch": 0.0016909744240118367, "eval_steps": 5, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0001690974424011837, "eval_loss": 11.938138961791992, "eval_runtime": 63.3507, "eval_samples_per_second": 39.305, "eval_steps_per_second": 19.653, "step": 1 }, { "epoch": 0.000507292327203551, "grad_norm": 0.1530068963766098, "learning_rate": 0.00012, "loss": 11.9354, "step": 3 }, { "epoch": 0.0008454872120059184, "eval_loss": 11.936454772949219, "eval_runtime": 63.4003, "eval_samples_per_second": 39.274, "eval_steps_per_second": 19.637, "step": 5 }, { "epoch": 0.001014584654407102, "grad_norm": 0.2502671182155609, "learning_rate": 0.0001992114701314478, "loss": 11.9346, "step": 6 }, { "epoch": 0.0015218769816106531, "grad_norm": 0.17158111929893494, "learning_rate": 0.00018763066800438636, "loss": 11.9326, "step": 9 }, { "epoch": 0.0016909744240118367, "eval_loss": 11.931650161743164, "eval_runtime": 63.4677, "eval_samples_per_second": 39.233, "eval_steps_per_second": 19.616, "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": 3110308282368.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }