{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.4230565838180857, "eval_steps": 10, "global_step": 50, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.008461131676361713, "eval_loss": NaN, "eval_runtime": 16.3579, "eval_samples_per_second": 48.723, "eval_steps_per_second": 6.113, "step": 1 }, { "epoch": 0.08461131676361713, "grad_norm": NaN, "learning_rate": 9.330127018922194e-05, "loss": 0.0, "step": 10 }, { "epoch": 0.08461131676361713, "eval_loss": NaN, "eval_runtime": 15.2988, "eval_samples_per_second": 52.095, "eval_steps_per_second": 6.536, "step": 10 }, { "epoch": 0.16922263352723427, "grad_norm": NaN, "learning_rate": 6.91341716182545e-05, "loss": 0.0, "step": 20 }, { "epoch": 0.16922263352723427, "eval_loss": NaN, "eval_runtime": 15.2436, "eval_samples_per_second": 52.284, "eval_steps_per_second": 6.56, "step": 20 }, { "epoch": 0.25383395029085143, "grad_norm": NaN, "learning_rate": 3.705904774487396e-05, "loss": 0.0, "step": 30 }, { "epoch": 0.25383395029085143, "eval_loss": NaN, "eval_runtime": 15.2213, "eval_samples_per_second": 52.361, "eval_steps_per_second": 6.57, "step": 30 }, { "epoch": 0.33844526705446853, "grad_norm": NaN, "learning_rate": 1.0332332985438248e-05, "loss": 0.0, "step": 40 }, { "epoch": 0.33844526705446853, "eval_loss": NaN, "eval_runtime": 15.2962, "eval_samples_per_second": 52.104, "eval_steps_per_second": 6.538, "step": 40 }, { "epoch": 0.4230565838180857, "grad_norm": NaN, "learning_rate": 0.0, "loss": 0.0, "step": 50 }, { "epoch": 0.4230565838180857, "eval_loss": NaN, "eval_runtime": 15.2148, "eval_samples_per_second": 52.383, "eval_steps_per_second": 6.573, "step": 50 } ], "logging_steps": 10, "max_steps": 50, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 13, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 2.59438606811136e+16, "train_batch_size": 8, "trial_name": null, "trial_params": null }