| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 20, | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 1.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 1e-06, | |
| "loss": 0.0, | |
| "num_tokens": 2080.0, | |
| "reward": 2.215920627117157, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 2.1505000591278076, | |
| "rewards/concensus_correctness_reward_func/std": 0.9416583180427551, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.06542056053876877, | |
| "rewards/question_recreation_reward_func/std": 0.0, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 2.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 9.729086208503173e-07, | |
| "loss": 0.0, | |
| "num_tokens": 4160.0, | |
| "reward": 0.6780470022931695, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 0.6675000190734863, | |
| "rewards/concensus_correctness_reward_func/std": 0.7707626819610596, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.010547009063884616, | |
| "rewards/question_recreation_reward_func/std": 0.002309401286765933, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 3.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 8.945702546981968e-07, | |
| "loss": 0.0, | |
| "num_tokens": 6240.0, | |
| "reward": 2.3659205436706543, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 2.2985000610351562, | |
| "rewards/concensus_correctness_reward_func/std": 0.7707626819610596, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.06742056063376367, | |
| "rewards/question_recreation_reward_func/std": 0.002309401286765933, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 4.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 7.734740790612136e-07, | |
| "loss": 0.0, | |
| "num_tokens": 8320.0, | |
| "reward": 3.709144711494446, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 3.5920000076293945, | |
| "rewards/concensus_correctness_reward_func/std": 0.0011546856258064508, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.11714470013976097, | |
| "rewards/question_recreation_reward_func/std": 0.0, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 5.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 6.227427435703995e-07, | |
| "loss": 0.0, | |
| "num_tokens": 10400.0, | |
| "reward": 0.6780470022931695, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 0.6675000190734863, | |
| "rewards/concensus_correctness_reward_func/std": 0.7707626819610596, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.010547009063884616, | |
| "rewards/question_recreation_reward_func/std": 0.002309401286765933, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 6.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 4.5871032726383385e-07, | |
| "loss": 0.0, | |
| "num_tokens": 12480.0, | |
| "reward": 3.709144711494446, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 3.5920000076293945, | |
| "rewards/concensus_correctness_reward_func/std": 0.0011546856258064508, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.11714470013976097, | |
| "rewards/question_recreation_reward_func/std": 0.0, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 7.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 2.9915228767351535e-07, | |
| "loss": 0.0, | |
| "num_tokens": 14560.0, | |
| "reward": 0.889467597939074, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 0.815500020980835, | |
| "rewards/concensus_correctness_reward_func/std": 0.9416583180427551, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.07396756950765848, | |
| "rewards/question_recreation_reward_func/std": 0.0, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 8.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 1.6135921418712955e-07, | |
| "loss": 0.0, | |
| "num_tokens": 16640.0, | |
| "reward": 1.7049675593152642, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 1.63100004196167, | |
| "rewards/concensus_correctness_reward_func/std": 0.0, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.07396756950765848, | |
| "rewards/question_recreation_reward_func/std": 0.0, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 9.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 6.026312439675551e-08, | |
| "loss": 0.0, | |
| "num_tokens": 18720.0, | |
| "reward": 0.889467597939074, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 0.815500020980835, | |
| "rewards/concensus_correctness_reward_func/std": 0.9416583180427551, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.07396756950765848, | |
| "rewards/question_recreation_reward_func/std": 0.0, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "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": 4.0, | |
| "completions/max_terminated_length": 4.0, | |
| "completions/mean_length": 4.0, | |
| "completions/mean_terminated_length": 4.0, | |
| "completions/min_length": 4.0, | |
| "completions/min_terminated_length": 4.0, | |
| "epoch": 10.0, | |
| "frac_reward_zero_std": 1.0, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 6.819348298638839e-09, | |
| "loss": 0.0, | |
| "num_tokens": 20800.0, | |
| "reward": 2.021271170116961, | |
| "reward_std": 0.0, | |
| "rewards/concensus_correctness_reward_func/mean": 1.9609999656677246, | |
| "rewards/concensus_correctness_reward_func/std": 0.0011546856258064508, | |
| "rewards/consensus_reward_func/mean": 0.0, | |
| "rewards/consensus_reward_func/std": 0.0, | |
| "rewards/cumulative_reward_2/mean": 0.0, | |
| "rewards/cumulative_reward_2/std": 0.0, | |
| "rewards/final_correctness_reward_func/mean": 0.0, | |
| "rewards/final_correctness_reward_func/std": 0.0, | |
| "rewards/question_recreation_reward_func/mean": 0.060271148569881916, | |
| "rewards/question_recreation_reward_func/std": 0.0, | |
| "rewards/soft_format_reward_func/mean": 0.0, | |
| "rewards/soft_format_reward_func/std": 0.0, | |
| "rewards/strict_format_reward_func/mean": 0.0, | |
| "rewards/strict_format_reward_func/std": 0.0, | |
| "rewards/xmlcount_reward_func/mean": 0.0, | |
| "rewards/xmlcount_reward_func/std": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 0.0, | |
| "train_runtime": 1185.7286, | |
| "train_samples_per_second": 0.067, | |
| "train_steps_per_second": 0.017 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 20800, | |
| "num_train_epochs": 10, | |
| "save_steps": 25, | |
| "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 | |
| } | |