{ "best_metric": NaN, "best_model_checkpoint": "miner_id_24/checkpoint-10", "epoch": 0.0001293259531322746, "eval_steps": 5, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.2932595313227458e-05, "eval_loss": NaN, "eval_runtime": 4511.4031, "eval_samples_per_second": 7.217, "eval_steps_per_second": 3.608, "step": 1 }, { "epoch": 3.8797785939682375e-05, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 3 }, { "epoch": 6.46629765661373e-05, "eval_loss": NaN, "eval_runtime": 4495.0464, "eval_samples_per_second": 7.243, "eval_steps_per_second": 3.622, "step": 5 }, { "epoch": 7.759557187936475e-05, "grad_norm": NaN, "learning_rate": 0.0001992114701314478, "loss": 0.0, "step": 6 }, { "epoch": 0.00011639335781904713, "grad_norm": NaN, "learning_rate": 0.00018763066800438636, "loss": 0.0, "step": 9 }, { "epoch": 0.0001293259531322746, "eval_loss": NaN, "eval_runtime": 4524.3529, "eval_samples_per_second": 7.196, "eval_steps_per_second": 3.598, "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": 3267293022781440.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }