{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 1e-05, "eval_steps": 500, "global_step": 1, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 374.0, "completions/max_terminated_length": 374.0, "completions/mean_length": 270.5625, "completions/mean_terminated_length": 270.5625, "completions/min_length": 212.0, "completions/min_terminated_length": 212.0, "entropy": 0.3388800360262394, "epoch": 1e-05, "frac_reward_zero_std": 0.875, "grad_norm": 1.3853623867034912, "kl": 0.0, "learning_rate": 0.0, "loss": -0.0033, "num_tokens": 26491.0, "reward": 0.06875000149011612, "reward_std": 0.012500000186264515, "rewards/rollout_reward_func/mean": 0.06875000149011612, "rewards/rollout_reward_func/std": 0.20703357458114624, "sampling/importance_sampling_ratio/max": 2.543839931488037, "sampling/importance_sampling_ratio/mean": 1.0151900053024292, "sampling/importance_sampling_ratio/min": 0.5303558707237244, "sampling/sampling_logp_difference/max": 1.07841157913208, "sampling/sampling_logp_difference/mean": 0.05954907462000847, "step": 1, "step_time": 6.106758965000154 } ], "logging_steps": 1.0, "max_steps": 200000, "num_input_tokens_seen": 26491, "num_train_epochs": 2, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }