{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 4e-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": 9190.0, "completions/max_terminated_length": 9190.0, "completions/mean_length": 8091.34375, "completions/mean_terminated_length": 8091.34375, "completions/min_length": 1965.0, "completions/min_terminated_length": 1965.0, "entropy": 0.09611401055008173, "epoch": 4e-05, "frac_reward_zero_std": 0.0, "grad_norm": 1.3341071605682373, "kl": 0.0, "learning_rate": 0.0, "loss": -0.0359, "num_tokens": 285973.0, "reward": -0.5604491233825684, "reward_std": 0.30373895168304443, "rewards/rollout_reward_func/mean": -0.5604491233825684, "rewards/rollout_reward_func/std": 0.393892765045166, "sampling/importance_sampling_ratio/max": 1.5303804874420166, "sampling/importance_sampling_ratio/mean": 1.0008273124694824, "sampling/importance_sampling_ratio/min": 0.5223565101623535, "sampling/sampling_logp_difference/max": 0.6494050025939941, "sampling/sampling_logp_difference/mean": 0.009981835260987282, "step": 1, "step_time": 98.97818489699921 } ], "logging_steps": 1.0, "max_steps": 400, "num_input_tokens_seen": 285973, "num_train_epochs": 1, "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": 1, "trial_name": null, "trial_params": null }