{ "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": 114.5, "completions/max_terminated_length": 114.5, "completions/mean_length": 111.25, "completions/mean_terminated_length": 111.25, "completions/min_length": 108.0, "completions/min_terminated_length": 108.0, "epoch": 0.5714285714285714, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.666666666666666e-07, "loss": 0.0, "num_tokens": 2938.0, "reward": 7.67693305015564, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9179999828338623, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 125.5, "completions/max_terminated_length": 125.5, "completions/mean_length": 121.66666412353516, "completions/mean_terminated_length": 121.66666412353516, "completions/min_length": 116.5, "completions/min_terminated_length": 116.5, "epoch": 1.2857142857142856, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 8e-07, "loss": 0.0, "num_tokens": 5954.0, "reward": 6.134842395782471, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.4668421894311905, "rewards/question_recreation_reward_func/std": 0.5008190274238586, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 129.0, "completions/max_terminated_length": 129.0, "completions/mean_length": 113.91666793823242, "completions/mean_terminated_length": 113.91666793823242, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 1.8571428571428572, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.96430995261912e-07, "loss": 0.0, "num_tokens": 8902.0, "reward": 6.701709508895874, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.5357092022895813, "rewards/question_recreation_reward_func/std": 0.5361168682575226, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 109.5, "completions/max_terminated_length": 109.5, "completions/mean_length": 102.75, "completions/mean_terminated_length": 102.75, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 2.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.857876700217506e-07, "loss": 0.0, "num_tokens": 11772.0, "reward": 6.697775840759277, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 125.5, "completions/max_terminated_length": 125.5, "completions/mean_length": 114.33333206176758, "completions/mean_terminated_length": 114.33333206176758, "completions/min_length": 104.5, "completions/min_terminated_length": 104.5, "epoch": 3.2857142857142856, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.682599546705714e-07, "loss": 0.0, "num_tokens": 14740.0, "reward": 6.227731227874756, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.5617313385009766, "rewards/question_recreation_reward_func/std": 0.5060690939426422, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 118.0, "completions/max_terminated_length": 118.0, "completions/mean_length": 111.58333206176758, "completions/mean_terminated_length": 111.58333206176758, "completions/min_length": 108.0, "completions/min_terminated_length": 108.0, "epoch": 3.857142857142857, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.441606317040557e-07, "loss": 0.0, "num_tokens": 17692.0, "reward": 7.679747819900513, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.916000008583069, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 120.5, "completions/max_terminated_length": 120.5, "completions/mean_length": 112.5, "completions/mean_terminated_length": 112.5, "completions/min_length": 104.5, "completions/min_terminated_length": 104.5, "epoch": 4.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.139197541114644e-07, "loss": 0.0, "num_tokens": 20640.0, "reward": 6.190184593200684, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.5241846442222595, "rewards/question_recreation_reward_func/std": 0.5494242608547211, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 113.0, "completions/max_terminated_length": 113.0, "completions/mean_length": 105.83333206176758, "completions/mean_terminated_length": 105.83333206176758, "completions/min_length": 97.5, "completions/min_terminated_length": 97.5, "epoch": 5.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 6.780769710698569e-07, "loss": 0.0, "num_tokens": 23530.0, "reward": 6.729287624359131, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 125.5, "completions/max_terminated_length": 125.5, "completions/mean_length": 121.16666793823242, "completions/mean_terminated_length": 121.16666793823242, "completions/min_length": 118.0, "completions/min_terminated_length": 118.0, "epoch": 5.857142857142857, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 6.37271897891742e-07, "loss": 0.0, "num_tokens": 26552.0, "reward": 6.184166669845581, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9140000343322754, "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": 0.5201666951179504, "rewards/question_recreation_reward_func/std": 0.5540637671947479, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 129.0, "completions/max_terminated_length": 129.0, "completions/mean_length": 121.0, "completions/mean_terminated_length": 121.0, "completions/min_length": 113.0, "completions/min_terminated_length": 113.0, "epoch": 6.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 5.922327020746735e-07, "loss": 0.0, "num_tokens": 29568.0, "reward": 6.147816181182861, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.4818165749311447, "rewards/question_recreation_reward_func/std": 0.48352810740470886, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 122.0, "completions/max_terminated_length": 122.0, "completions/mean_length": 117.75, "completions/mean_terminated_length": 117.75, "completions/min_length": 113.0, "completions/min_terminated_length": 113.0, "epoch": 7.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 5.437631091350051e-07, "loss": 0.0, "num_tokens": 32556.0, "reward": 6.695430040359497, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.5294299721717834, "rewards/question_recreation_reward_func/std": 0.543367475271225, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 118.0, "completions/max_terminated_length": 118.0, "completions/mean_length": 111.58333206176758, "completions/mean_terminated_length": 111.58333206176758, "completions/min_length": 108.0, "completions/min_terminated_length": 108.0, "epoch": 7.857142857142857, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 4.927280601070113e-07, "loss": 0.0, "num_tokens": 35508.0, "reward": 7.68305778503418, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9179999828338623, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 125.5, "completions/max_terminated_length": 125.5, "completions/mean_length": 111.5, "completions/mean_terminated_length": 111.5, "completions/min_length": 97.5, "completions/min_terminated_length": 97.5, "epoch": 8.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 4.400382766496394e-07, "loss": 0.0, "num_tokens": 38448.0, "reward": 6.215600967407227, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.5476008653640747, "rewards/question_recreation_reward_func/std": 0.5223855376243591, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 116.5, "completions/max_terminated_length": 116.5, "completions/mean_length": 107.66666793823242, "completions/mean_terminated_length": 107.66666793823242, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 9.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 3.866340091969303e-07, "loss": 0.0, "num_tokens": 41346.0, "reward": 8.184330940246582, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.916000008583069, "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": 0.5183308720588684, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 121.5, "completions/max_terminated_length": 121.5, "completions/mean_length": 114.91666793823242, "completions/mean_terminated_length": 114.91666793823242, "completions/min_length": 109.5, "completions/min_terminated_length": 109.5, "epoch": 9.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 3.33468258166748e-07, "loss": 0.0, "num_tokens": 44318.0, "reward": 6.189554929733276, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9140000343322754, "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": 0.5255548357963562, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 113.0, "completions/max_terminated_length": 113.0, "completions/mean_length": 104.5, "completions/mean_terminated_length": 104.5, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 10.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.8148976764574644e-07, "loss": 0.0, "num_tokens": 47202.0, "reward": 6.747161626815796, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.579161524772644, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 133.0, "completions/max_terminated_length": 133.0, "completions/mean_length": 125.58333206176758, "completions/mean_terminated_length": 125.58333206176758, "completions/min_length": 114.5, "completions/min_terminated_length": 114.5, "epoch": 11.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.3162609502866607e-07, "loss": 0.0, "num_tokens": 50240.0, "reward": 6.091987371444702, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.4239872694015503, "rewards/question_recreation_reward_func/std": 0.43548497557640076, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 118.0, "completions/max_terminated_length": 118.0, "completions/mean_length": 109.25, "completions/mean_terminated_length": 109.25, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 11.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 1.8476705873465095e-07, "loss": 0.0, "num_tokens": 53144.0, "reward": 6.189610004425049, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.5236097574234009, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 120.5, "completions/max_terminated_length": 120.5, "completions/mean_length": 110.75, "completions/mean_terminated_length": 110.75, "completions/min_length": 101.0, "completions/min_terminated_length": 101.0, "epoch": 12.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 1.41748859376479e-07, "loss": 0.0, "num_tokens": 56078.0, "reward": 6.197602987289429, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.5296030640602112, "rewards/question_recreation_reward_func/std": 0.5431675910949707, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 125.0, "completions/max_terminated_length": 125.0, "completions/mean_length": 119.16666793823242, "completions/mean_terminated_length": 119.16666793823242, "completions/min_length": 111.0, "completions/min_terminated_length": 111.0, "epoch": 13.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 1.0333915774081696e-07, "loss": 0.0, "num_tokens": 59070.0, "reward": 6.176448345184326, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9140000343322754, "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": 0.5124481320381165, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 120.5, "completions/max_terminated_length": 120.5, "completions/mean_length": 106.16666793823242, "completions/mean_terminated_length": 106.16666793823242, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 13.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 7.022337586329597e-08, "loss": 0.0, "num_tokens": 61984.0, "reward": 7.68181848526001, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 2.9179999828338623, "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": 0.5138185024261475, "rewards/question_recreation_reward_func/std": 0.561394065618515, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 122.0, "completions/max_terminated_length": 122.0, "completions/mean_length": 115.0, "completions/mean_terminated_length": 115.0, "completions/min_length": 108.0, "completions/min_terminated_length": 108.0, "epoch": 14.571428571428571, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 4.299246565604755e-08, "loss": 0.0, "num_tokens": 64952.0, "reward": 6.64246392250061, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.47646407783031464, "rewards/question_recreation_reward_func/std": 0.48970863223075867, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 121.5, "completions/max_terminated_length": 121.5, "completions/mean_length": 106.08333206176758, "completions/mean_terminated_length": 106.08333206176758, "completions/min_length": 97.5, "completions/min_terminated_length": 97.5, "epoch": 15.285714285714286, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 2.213236335683253e-08, "loss": 0.0, "num_tokens": 67876.0, "reward": 6.259791612625122, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.591791570186615, "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": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 129.0, "completions/max_terminated_length": 129.0, "completions/mean_length": 113.91666793823242, "completions/mean_terminated_length": 113.91666793823242, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 15.857142857142858, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 8.0153179853653e-09, "loss": 0.0, "num_tokens": 70824.0, "reward": 6.701709508895874, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9160000085830688, "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": 0.5357092022895813, "rewards/question_recreation_reward_func/std": 0.5361168682575226, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "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": 120.5, "completions/max_terminated_length": 120.5, "completions/mean_length": 108.25, "completions/mean_terminated_length": 108.25, "completions/min_length": 96.0, "completions/min_terminated_length": 96.0, "epoch": 16.571428571428573, "grad_norm": 0.0, "kl": 0.0, "learning_rate": 8.932485507387344e-10, "loss": 0.0, "num_tokens": 73738.0, "reward": 6.639105796813965, "reward_std": 0.0, "rewards/concensus_correctness_reward_func/mean": 1.9179999828338623, "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": 0.47110553085803986, "rewards/question_recreation_reward_func/std": 0.4958961009979248, "rewards/soft_format_reward_func/mean": 0.0, "rewards/soft_format_reward_func/std": 0.0, "rewards/strict_format_reward_func/mean": 0.5, "rewards/strict_format_reward_func/std": 0.0, "rewards/xmlcount_reward_func/mean": 1.25, "rewards/xmlcount_reward_func/std": 0.0, "step": 50 }, { "epoch": 16.571428571428573, "step": 50, "total_flos": 0.0, "train_loss": 0.0, "train_runtime": 1743.1391, "train_samples_per_second": 0.115, "train_steps_per_second": 0.029 } ], "logging_steps": 2, "max_steps": 50, "num_input_tokens_seen": 73738, "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 }