{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 16.571428571428573, "eval_steps": 500, "global_step": 50, "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": 147.5, "completions/max_terminated_length": 147.5, "completions/mean_length": 135.125, "completions/mean_terminated_length": 135.125, "completions/min_length": 120.5, "completions/min_terminated_length": 120.5, "epoch": 0.5714285714285714, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.666666666666666e-07, "loss": 0.0, "num_tokens": 3129.0, "reward": 6.070932626724243, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5089328289031982, "rewards/question_recreation_reward_func/std": 0.5670355260372162, "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": 1.1380000114440918, "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": 148.5, "completions/max_terminated_length": 148.5, "completions/mean_length": 144.45833587646484, "completions/mean_terminated_length": 144.45833587646484, "completions/min_length": 139.0, "completions/min_terminated_length": 139.0, "epoch": 1.2857142857142856, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 8e-07, "loss": 0.0, "num_tokens": 6316.0, "reward": 5.576179265975952, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5141791701316833, "rewards/question_recreation_reward_func/std": 0.5554790496826172, "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": 1.1380000114440918, "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": 161.0, "completions/max_terminated_length": 161.0, "completions/mean_length": 141.16666412353516, "completions/mean_terminated_length": 141.16666412353516, "completions/min_length": 120.5, "completions/min_terminated_length": 120.5, "epoch": 1.8571428571428572, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.96430995261912e-07, "loss": 0.0, "num_tokens": 9464.0, "reward": 5.596338510513306, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.534338653087616, "rewards/question_recreation_reward_func/std": 0.5376994013786316, "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": 1.1380000114440918, "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": 134.5, "completions/max_terminated_length": 134.5, "completions/mean_length": 121.625, "completions/mean_terminated_length": 121.625, "completions/min_length": 115.5, "completions/min_terminated_length": 115.5, "epoch": 2.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.857876700217506e-07, "loss": 0.0, "num_tokens": 12485.0, "reward": 7.093775987625122, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5317758321762085, "rewards/question_recreation_reward_func/std": 0.5406587421894073, "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": 1.1380000114440918, "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": 148.0, "completions/max_terminated_length": 148.0, "completions/mean_length": 133.0, "completions/mean_terminated_length": 133.0, "completions/min_length": 116.0, "completions/min_terminated_length": 116.0, "epoch": 3.2857142857142856, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.682599546705714e-07, "loss": 0.0, "num_tokens": 15605.0, "reward": 5.623394966125488, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5613951683044434, "rewards/question_recreation_reward_func/std": 0.5064572691917419, "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": 1.1380000114440918, "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": 139.0, "completions/max_terminated_length": 139.0, "completions/mean_length": 129.04166412353516, "completions/mean_terminated_length": 129.04166412353516, "completions/min_length": 122.5, "completions/min_terminated_length": 122.5, "epoch": 3.857142857142857, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.441606317040557e-07, "loss": 0.0, "num_tokens": 18689.0, "reward": 7.075747728347778, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5137476325035095, "rewards/question_recreation_reward_func/std": 0.5614758431911469, "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": 1.1380000114440918, "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": 144.0, "completions/max_terminated_length": 144.0, "completions/mean_length": 131.75, "completions/mean_terminated_length": 131.75, "completions/min_length": 121.5, "completions/min_terminated_length": 121.5, "epoch": 4.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.139197541114644e-07, "loss": 0.0, "num_tokens": 21791.0, "reward": 5.585848331451416, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5238484740257263, "rewards/question_recreation_reward_func/std": 0.5498124361038208, "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": 1.1380000114440918, "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": 133.5, "completions/max_terminated_length": 133.5, "completions/mean_length": 125.125, "completions/mean_terminated_length": 125.125, "completions/min_length": 115.5, "completions/min_terminated_length": 115.5, "epoch": 5.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 6.780769710698569e-07, "loss": 0.0, "num_tokens": 24827.0, "reward": 6.1232874393463135, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5612876415252686, "rewards/question_recreation_reward_func/std": 0.5065814107656479, "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": 1.1380000114440918, "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": 148.0, "completions/max_terminated_length": 148.0, "completions/mean_length": 137.875, "completions/mean_terminated_length": 137.875, "completions/min_length": 127.0, "completions/min_terminated_length": 127.0, "epoch": 5.857142857142857, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 6.37271897891742e-07, "loss": 0.0, "num_tokens": 27995.0, "reward": 7.081830263137817, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5198305249214172, "rewards/question_recreation_reward_func/std": 0.5544519424438477, "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": 1.1380000114440918, "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": 155.5, "completions/max_terminated_length": 155.5, "completions/mean_length": 142.25, "completions/mean_terminated_length": 142.25, "completions/min_length": 132.0, "completions/min_terminated_length": 132.0, "epoch": 6.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 5.922327020746735e-07, "loss": 0.0, "num_tokens": 31181.0, "reward": 5.591153383255005, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5291535258293152, "rewards/question_recreation_reward_func/std": 0.5381881296634674, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 20 }, { "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": 152.0, "completions/max_terminated_length": 152.0, "completions/mean_length": 142.29166412353516, "completions/mean_terminated_length": 142.29166412353516, "completions/min_length": 128.5, "completions/min_terminated_length": 128.5, "epoch": 7.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 5.437631091350051e-07, "loss": 0.0, "num_tokens": 34351.0, "reward": 7.0908520221710205, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5288519263267517, "rewards/question_recreation_reward_func/std": 0.5440349578857422, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 22 }, { "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": 146.5, "completions/max_terminated_length": 146.5, "completions/mean_length": 127.70833206176758, "completions/mean_terminated_length": 127.70833206176758, "completions/min_length": 120.5, "completions/min_terminated_length": 120.5, "epoch": 7.857142857142857, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 4.927280601070113e-07, "loss": 0.0, "num_tokens": 37429.0, "reward": 6.077057361602783, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5150575637817383, "rewards/question_recreation_reward_func/std": 0.5599632859230042, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 24 }, { "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": 154.0, "completions/max_terminated_length": 154.0, "completions/mean_length": 134.0, "completions/mean_terminated_length": 134.0, "completions/min_length": 108.5, "completions/min_terminated_length": 108.5, "epoch": 8.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 4.400382766496394e-07, "loss": 0.0, "num_tokens": 40549.0, "reward": 6.108230352401733, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5462303161621094, "rewards/question_recreation_reward_func/std": 0.5239680707454681, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 26 }, { "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": 145.5, "completions/max_terminated_length": 145.5, "completions/mean_length": 128.25, "completions/mean_terminated_length": 128.25, "completions/min_length": 109.0, "completions/min_terminated_length": 109.0, "epoch": 9.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 3.866340091969303e-07, "loss": 0.0, "num_tokens": 43619.0, "reward": 5.5803306102752686, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5183308720588684, "rewards/question_recreation_reward_func/std": 0.5561836063861847, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 28 }, { "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": 138.0, "completions/max_terminated_length": 138.0, "completions/mean_length": 132.70833587646484, "completions/mean_terminated_length": 132.70833587646484, "completions/min_length": 126.5, "completions/min_terminated_length": 126.5, "epoch": 9.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 3.33468258166748e-07, "loss": 0.0, "num_tokens": 46738.0, "reward": 7.087554454803467, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5255548357963562, "rewards/question_recreation_reward_func/std": 0.5478420853614807, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 30 }, { "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": 139.5, "completions/max_terminated_length": 139.5, "completions/mean_length": 127.25, "completions/mean_terminated_length": 127.25, "completions/min_length": 109.0, "completions/min_terminated_length": 109.0, "epoch": 10.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.8148976764574644e-07, "loss": 0.0, "num_tokens": 49804.0, "reward": 5.6411614418029785, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.579161524772644, "rewards/question_recreation_reward_func/std": 0.48594246804714203, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 32 }, { "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": 161.5, "completions/max_terminated_length": 161.5, "completions/mean_length": 146.04166412353516, "completions/mean_terminated_length": 146.04166412353516, "completions/min_length": 127.0, "completions/min_terminated_length": 127.0, "epoch": 11.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.3162609502866607e-07, "loss": 0.0, "num_tokens": 53020.0, "reward": 6.080661296844482, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.518661230802536, "rewards/question_recreation_reward_func/std": 0.5448050200939178, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 34 }, { "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": 134.0, "completions/max_terminated_length": 134.0, "completions/mean_length": 125.20833206176758, "completions/mean_terminated_length": 125.20833206176758, "completions/min_length": 115.0, "completions/min_terminated_length": 115.0, "epoch": 11.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 1.8476705873465095e-07, "loss": 0.0, "num_tokens": 56199.0, "reward": 7.085609674453735, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5236097574234009, "rewards/question_recreation_reward_func/std": 0.5500880181789398, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 36 }, { "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": 142.5, "completions/max_terminated_length": 142.5, "completions/mean_length": 132.75, "completions/mean_terminated_length": 132.75, "completions/min_length": 124.5, "completions/min_terminated_length": 124.5, "epoch": 12.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 1.41748859376479e-07, "loss": 0.0, "num_tokens": 59309.0, "reward": 5.591024875640869, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5290250182151794, "rewards/question_recreation_reward_func/std": 0.5438350737094879, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 38 }, { "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": 141.0, "completions/max_terminated_length": 141.0, "completions/mean_length": 133.58333587646484, "completions/mean_terminated_length": 133.58333587646484, "completions/min_length": 127.0, "completions/min_terminated_length": 127.0, "epoch": 13.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 1.0333915774081696e-07, "loss": 0.0, "num_tokens": 62434.0, "reward": 7.574448108673096, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 1.0, "rewards/final_correctness_reward_func/std": 1.154700517654419, "rewards/question_recreation_reward_func/mean": 0.5124481320381165, "rewards/question_recreation_reward_func/std": 0.5629763901233673, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 40 }, { "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": 140.5, "completions/max_terminated_length": 140.5, "completions/mean_length": 125.29166412353516, "completions/mean_terminated_length": 125.29166412353516, "completions/min_length": 115.0, "completions/min_terminated_length": 115.0, "epoch": 13.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.022337586329597e-08, "loss": 0.0, "num_tokens": 65500.0, "reward": 5.575525760650635, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5135260820388794, "rewards/question_recreation_reward_func/std": 0.5617317259311676, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 42 }, { "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": 140.5, "completions/max_terminated_length": 140.5, "completions/mean_length": 133.625, "completions/mean_terminated_length": 133.625, "completions/min_length": 126.5, "completions/min_terminated_length": 126.5, "epoch": 14.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 4.299246565604755e-08, "loss": 0.0, "num_tokens": 68617.0, "reward": 7.085801124572754, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.5238010287284851, "rewards/question_recreation_reward_func/std": 0.5443686544895172, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 44 }, { "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": 136.5, "completions/max_terminated_length": 136.5, "completions/mean_length": 124.83333206176758, "completions/mean_terminated_length": 124.83333206176758, "completions/min_length": 114.0, "completions/min_terminated_length": 114.0, "epoch": 15.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.213236335683253e-08, "loss": 0.0, "num_tokens": 71685.0, "reward": 6.153791427612305, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5, "rewards/final_correctness_reward_func/std": 0.5773502588272095, "rewards/question_recreation_reward_func/mean": 0.591791570186615, "rewards/question_recreation_reward_func/std": 0.47135844826698303, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 46 }, { "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": 156.0, "completions/max_terminated_length": 156.0, "completions/mean_length": 137.25, "completions/mean_terminated_length": 137.25, "completions/min_length": 115.5, "completions/min_terminated_length": 115.5, "epoch": 15.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 8.0153179853653e-09, "loss": 0.0, "num_tokens": 74824.0, "reward": 5.596338510513306, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.534338653087616, "rewards/question_recreation_reward_func/std": 0.5376994013786316, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 48 }, { "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": 144.0, "completions/max_terminated_length": 144.0, "completions/mean_length": 126.625, "completions/mean_terminated_length": 126.625, "completions/min_length": 108.5, "completions/min_terminated_length": 108.5, "epoch": 16.571428571428573, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 8.932485507387344e-10, "loss": 0.0, "num_tokens": 77885.0, "reward": 5.580442428588867, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9240000247955322, "rewards/concensus_correctness_reward_func/std": 0.0, "rewards/consensus_reward_func/mean": 2.0, "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": 0.5184425115585327, "rewards/question_recreation_reward_func/std": 0.550556093454361, "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": 1.1380000114440918, "rewards/xmlcount_reward_func/std": 0.0, "step": 50 }, { "epoch": 16.571428571428573, "step": 50, "total_flos": 0.0, "train_loss": 0.0, "train_runtime": 2413.8883, "train_samples_per_second": 0.083, "train_steps_per_second": 0.021 } ], "logging_steps": 2, "max_steps": 50, "num_input_tokens_seen": 77885, "num_train_epochs": 17, "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": 1, "trial_name": null, "trial_params": null }