{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7692307692307693,
  "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": 0.25,
      "completions/max_length": 448.0,
      "completions/max_terminated_length": 190.0,
      "completions/mean_length": 218.0,
      "completions/mean_terminated_length": 141.3333282470703,
      "completions/min_length": 91.0,
      "completions/min_terminated_length": 91.0,
      "epoch": 0.038461538461538464,
      "grad_norm": 9.404427528381348,
      "kl": 0.0,
      "learning_rate": 0.0,
      "loss": 0.2213,
      "num_tokens": 1384.0,
      "reward": 0.27538180351257324,
      "reward_std": 0.5752406120300293,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.3473817706108093,
      "rewards/question_recreation_reward_func/std": 0.03370636701583862,
      "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": -0.07199999690055847,
      "rewards/xmlcount_reward_func/std": 0.9370037317276001,
      "step": 1
    },
    {
      "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": 297.0,
      "completions/max_terminated_length": 297.0,
      "completions/mean_length": 123.75,
      "completions/mean_terminated_length": 123.75,
      "completions/min_length": 39.0,
      "completions/min_terminated_length": 39.0,
      "epoch": 0.07692307692307693,
      "grad_norm": 13.685848236083984,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.1909,
      "num_tokens": 2391.0,
      "reward": 0.2730637192726135,
      "reward_std": 0.18166151642799377,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.023063700646162033,
      "rewards/question_recreation_reward_func/std": 0.009388059377670288,
      "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": 0.25,
      "rewards/xmlcount_reward_func/std": 0.20412415266036987,
      "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": 212.0,
      "completions/max_terminated_length": 212.0,
      "completions/mean_length": 170.0,
      "completions/mean_terminated_length": 170.0,
      "completions/min_length": 118.0,
      "completions/min_terminated_length": 118.0,
      "epoch": 0.11538461538461539,
      "grad_norm": 8.721884727478027,
      "kl": 0.0005104035590193234,
      "learning_rate": 9.931806517013612e-07,
      "loss": -0.1137,
      "num_tokens": 3583.0,
      "reward": 0.5263556241989136,
      "reward_std": 0.10577338933944702,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.08885563910007477,
      "rewards/question_recreation_reward_func/std": 0.022719671949744225,
      "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": 0.4375,
      "rewards/xmlcount_reward_func/std": 0.125,
      "step": 3
    },
    {
      "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": 221.0,
      "completions/max_terminated_length": 221.0,
      "completions/mean_length": 89.75,
      "completions/mean_terminated_length": 89.75,
      "completions/min_length": 35.0,
      "completions/min_terminated_length": 35.0,
      "epoch": 0.15384615384615385,
      "grad_norm": 18.96868324279785,
      "kl": 0.0063872426908346824,
      "learning_rate": 9.729086208503173e-07,
      "loss": 0.2846,
      "num_tokens": 4454.0,
      "reward": 0.8816813230514526,
      "reward_std": 0.8639886379241943,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.5,
      "rewards/final_correctness_reward_func/std": 1.0,
      "rewards/question_recreation_reward_func/mean": 0.13168135285377502,
      "rewards/question_recreation_reward_func/std": 0.041856855154037476,
      "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": 0.25,
      "rewards/xmlcount_reward_func/std": 0.20412415266036987,
      "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": 0.25,
      "completions/max_length": 448.0,
      "completions/max_terminated_length": 168.0,
      "completions/mean_length": 185.25,
      "completions/mean_terminated_length": 97.66666412353516,
      "completions/min_length": 56.0,
      "completions/min_terminated_length": 56.0,
      "epoch": 0.19230769230769232,
      "grad_norm": 11.014723777770996,
      "kl": 0.0015677963965572417,
      "learning_rate": 9.397368756032444e-07,
      "loss": 0.4359,
      "num_tokens": 5707.0,
      "reward": 0.01884479820728302,
      "reward_std": 0.5485635995864868,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.03259478509426117,
      "rewards/question_recreation_reward_func/std": 0.005219103768467903,
      "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": -0.013750001788139343,
      "rewards/xmlcount_reward_func/std": 0.6304149031639099,
      "step": 5
    },
    {
      "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": 241.0,
      "completions/max_terminated_length": 241.0,
      "completions/mean_length": 112.75,
      "completions/mean_terminated_length": 112.75,
      "completions/min_length": 44.0,
      "completions/min_terminated_length": 44.0,
      "epoch": 0.23076923076923078,
      "grad_norm": 16.488595962524414,
      "kl": 0.013569895178079605,
      "learning_rate": 8.945702546981968e-07,
      "loss": 0.161,
      "num_tokens": 6670.0,
      "reward": 0.5298745632171631,
      "reward_std": 0.11173999309539795,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.03187461569905281,
      "rewards/question_recreation_reward_func/std": 0.011758098378777504,
      "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": 0.49799996614456177,
      "rewards/xmlcount_reward_func/std": 0.21535088121891022,
      "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": 0.25,
      "completions/max_length": 448.0,
      "completions/max_terminated_length": 93.0,
      "completions/mean_length": 173.25,
      "completions/mean_terminated_length": 81.66666412353516,
      "completions/min_length": 74.0,
      "completions/min_terminated_length": 74.0,
      "epoch": 0.2692307692307692,
      "grad_norm": 17.747295379638672,
      "kl": 0.05565625987946987,
      "learning_rate": 8.386407858128706e-07,
      "loss": 0.2248,
      "num_tokens": 7875.0,
      "reward": 0.4869118928909302,
      "reward_std": 0.11581836640834808,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.04941185563802719,
      "rewards/question_recreation_reward_func/std": 0.04165700823068619,
      "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": 0.4375,
      "rewards/xmlcount_reward_func/std": 0.125,
      "step": 7
    },
    {
      "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": 223.0,
      "completions/max_terminated_length": 223.0,
      "completions/mean_length": 114.75,
      "completions/mean_terminated_length": 114.75,
      "completions/min_length": 40.0,
      "completions/min_terminated_length": 40.0,
      "epoch": 0.3076923076923077,
      "grad_norm": 12.994662284851074,
      "kl": 0.030461058020591736,
      "learning_rate": 7.734740790612136e-07,
      "loss": -0.0224,
      "num_tokens": 8846.0,
      "reward": 0.5597025752067566,
      "reward_std": 0.1636485457420349,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.2159525454044342,
      "rewards/question_recreation_reward_func/std": 0.05881807208061218,
      "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": 0.34375,
      "rewards/xmlcount_reward_func/std": 0.1875,
      "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": 0.25,
      "completions/max_length": 448.0,
      "completions/max_terminated_length": 95.0,
      "completions/mean_length": 159.75,
      "completions/mean_terminated_length": 63.66666793823242,
      "completions/min_length": 42.0,
      "completions/min_terminated_length": 42.0,
      "epoch": 0.34615384615384615,
      "grad_norm": 16.384445190429688,
      "kl": 0.02970386017113924,
      "learning_rate": 7.008477123264847e-07,
      "loss": 0.2749,
      "num_tokens": 9997.0,
      "reward": 0.43550199270248413,
      "reward_std": 0.09783899784088135,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.12300197780132294,
      "rewards/question_recreation_reward_func/std": 0.0868106335401535,
      "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": 0.3125,
      "rewards/xmlcount_reward_func/std": 0.23935678601264954,
      "step": 9
    },
    {
      "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": 55.0,
      "completions/max_terminated_length": 55.0,
      "completions/mean_length": 47.0,
      "completions/mean_terminated_length": 47.0,
      "completions/min_length": 29.0,
      "completions/min_terminated_length": 29.0,
      "epoch": 0.38461538461538464,
      "grad_norm": 19.250350952148438,
      "kl": 0.039656256791204214,
      "learning_rate": 6.227427435703995e-07,
      "loss": -0.1137,
      "num_tokens": 10697.0,
      "reward": 0.6099977493286133,
      "reward_std": 0.021727532148361206,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.23549777269363403,
      "rewards/question_recreation_reward_func/std": 0.02440648153424263,
      "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": 0.37450000643730164,
      "rewards/xmlcount_reward_func/std": 0.1437625288963318,
      "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": 57.0,
      "completions/max_terminated_length": 57.0,
      "completions/mean_length": 51.5,
      "completions/mean_terminated_length": 51.5,
      "completions/min_length": 39.0,
      "completions/min_terminated_length": 39.0,
      "epoch": 0.4230769230769231,
      "grad_norm": 18.542329788208008,
      "kl": 0.020436149090528488,
      "learning_rate": 5.412896727361662e-07,
      "loss": -0.0652,
      "num_tokens": 11415.0,
      "reward": 0.6023240089416504,
      "reward_std": 0.06527406722307205,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.0183239858597517,
      "rewards/question_recreation_reward_func/std": 0.008217486552894115,
      "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": 0.5839999914169312,
      "rewards/xmlcount_reward_func/std": 0.11785867810249329,
      "step": 11
    },
    {
      "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": 91.0,
      "completions/max_terminated_length": 91.0,
      "completions/mean_length": 71.25,
      "completions/mean_terminated_length": 71.25,
      "completions/min_length": 52.0,
      "completions/min_terminated_length": 52.0,
      "epoch": 0.46153846153846156,
      "grad_norm": 14.851263046264648,
      "kl": 0.04291308671236038,
      "learning_rate": 4.5871032726383385e-07,
      "loss": -0.1437,
      "num_tokens": 12212.0,
      "reward": 0.5428086519241333,
      "reward_std": 0.00478853564709425,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.04280860722064972,
      "rewards/question_recreation_reward_func/std": 0.005272616166621447,
      "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": 0.5,
      "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": 0.25,
      "completions/max_length": 448.0,
      "completions/max_terminated_length": 65.0,
      "completions/mean_length": 153.25,
      "completions/mean_terminated_length": 55.0,
      "completions/min_length": 45.0,
      "completions/min_terminated_length": 45.0,
      "epoch": 0.5,
      "grad_norm": 12.394680976867676,
      "kl": 0.022660875925794244,
      "learning_rate": 3.772572564296004e-07,
      "loss": 0.341,
      "num_tokens": 13337.0,
      "reward": 0.3977251648902893,
      "reward_std": 0.18016040325164795,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.022725149989128113,
      "rewards/question_recreation_reward_func/std": 0.01266756746917963,
      "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": 0.375,
      "rewards/xmlcount_reward_func/std": 0.25,
      "step": 13
    },
    {
      "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": 180.0,
      "completions/max_terminated_length": 180.0,
      "completions/mean_length": 92.0,
      "completions/mean_terminated_length": 92.0,
      "completions/min_length": 27.0,
      "completions/min_terminated_length": 27.0,
      "epoch": 0.5384615384615384,
      "grad_norm": 26.36524200439453,
      "kl": 0.0786033570766449,
      "learning_rate": 2.9915228767351535e-07,
      "loss": 0.0012,
      "num_tokens": 14217.0,
      "reward": 0.8583254814147949,
      "reward_std": 0.6752742528915405,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.5,
      "rewards/final_correctness_reward_func/std": 1.0,
      "rewards/question_recreation_reward_func/mean": 0.13957545161247253,
      "rewards/question_recreation_reward_func/std": 0.03446739539504051,
      "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": 0.21875,
      "rewards/xmlcount_reward_func/std": 0.2576940953731537,
      "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": 268.0,
      "completions/max_terminated_length": 268.0,
      "completions/mean_length": 189.25,
      "completions/mean_terminated_length": 189.25,
      "completions/min_length": 124.0,
      "completions/min_terminated_length": 124.0,
      "epoch": 0.5769230769230769,
      "grad_norm": 9.72214126586914,
      "kl": 0.059507149271667004,
      "learning_rate": 2.2652592093878665e-07,
      "loss": -0.0895,
      "num_tokens": 15486.0,
      "reward": 0.6740732789039612,
      "reward_std": 0.2836502194404602,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.29907330870628357,
      "rewards/question_recreation_reward_func/std": 0.14153823256492615,
      "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": 0.375,
      "rewards/xmlcount_reward_func/std": 0.25,
      "step": 15
    },
    {
      "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": 0.25,
      "completions/max_length": 448.0,
      "completions/max_terminated_length": 150.0,
      "completions/mean_length": 178.0,
      "completions/mean_terminated_length": 88.0,
      "completions/min_length": 53.0,
      "completions/min_terminated_length": 53.0,
      "epoch": 0.6153846153846154,
      "grad_norm": 12.116292953491211,
      "kl": 0.025627892464399338,
      "learning_rate": 1.6135921418712955e-07,
      "loss": 0.2017,
      "num_tokens": 16710.0,
      "reward": 0.37010395526885986,
      "reward_std": 0.09250971674919128,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.05760394036769867,
      "rewards/question_recreation_reward_func/std": 0.03707271069288254,
      "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": 0.3125,
      "rewards/xmlcount_reward_func/std": 0.23935678601264954,
      "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": 155.0,
      "completions/max_terminated_length": 155.0,
      "completions/mean_length": 92.5,
      "completions/mean_terminated_length": 92.5,
      "completions/min_length": 57.0,
      "completions/min_terminated_length": 57.0,
      "epoch": 0.6538461538461539,
      "grad_norm": 11.819451332092285,
      "kl": 0.04005518322810531,
      "learning_rate": 1.0542974530180327e-07,
      "loss": -0.1202,
      "num_tokens": 17592.0,
      "reward": 0.5551368594169617,
      "reward_std": 0.021570861339569092,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.05513682961463928,
      "rewards/question_recreation_reward_func/std": 0.03840121626853943,
      "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": 0.5,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 17
    },
    {
      "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": 293.0,
      "completions/max_terminated_length": 293.0,
      "completions/mean_length": 165.0,
      "completions/mean_terminated_length": 165.0,
      "completions/min_length": 87.0,
      "completions/min_terminated_length": 87.0,
      "epoch": 0.6923076923076923,
      "grad_norm": 12.008862495422363,
      "kl": 0.03604450775310397,
      "learning_rate": 6.026312439675551e-08,
      "loss": -0.1226,
      "num_tokens": 18764.0,
      "reward": 0.46445250511169434,
      "reward_std": 0.07633315026760101,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.026952533051371574,
      "rewards/question_recreation_reward_func/std": 0.02285909652709961,
      "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": 0.4375,
      "rewards/xmlcount_reward_func/std": 0.125,
      "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": 99.0,
      "completions/max_terminated_length": 99.0,
      "completions/mean_length": 67.25,
      "completions/mean_terminated_length": 67.25,
      "completions/min_length": 46.0,
      "completions/min_terminated_length": 46.0,
      "epoch": 0.7307692307692307,
      "grad_norm": 15.652201652526855,
      "kl": 0.011676187859848142,
      "learning_rate": 2.7091379149682682e-08,
      "loss": 0.1494,
      "num_tokens": 19545.0,
      "reward": 0.5532028675079346,
      "reward_std": 0.015393441542983055,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.05320284515619278,
      "rewards/question_recreation_reward_func/std": 0.026126131415367126,
      "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": 0.5,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 19
    },
    {
      "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": 0.25,
      "completions/max_length": 448.0,
      "completions/max_terminated_length": 55.0,
      "completions/mean_length": 139.25,
      "completions/mean_terminated_length": 36.33333206176758,
      "completions/min_length": 9.0,
      "completions/min_terminated_length": 9.0,
      "epoch": 0.7692307692307693,
      "grad_norm": 12.616737365722656,
      "kl": 0.01751775573939085,
      "learning_rate": 6.819348298638839e-09,
      "loss": -0.3033,
      "num_tokens": 20614.0,
      "reward": 0.372824102640152,
      "reward_std": 0.048663586378097534,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "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": 0.02907411754131317,
      "rewards/question_recreation_reward_func/std": 0.014660523273050785,
      "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": 0.34375,
      "rewards/xmlcount_reward_func/std": 0.1875,
      "step": 20
    },
    {
      "epoch": 0.7692307692307693,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 0.0696285206824541,
      "train_runtime": 6061.8527,
      "train_samples_per_second": 0.013,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 20614,
  "num_train_epochs": 1,
  "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
}