{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 19.666666666666668,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 38.0,
      "completions/max_terminated_length": 38.0,
      "completions/mean_length": 29.33333396911621,
      "completions/mean_terminated_length": 29.33333396911621,
      "completions/min_length": 24.0,
      "completions/min_terminated_length": 24.0,
      "epoch": 0.6666666666666666,
      "grad_norm": 11.15837574005127,
      "kl": 0.0,
      "learning_rate": 0.0,
      "loss": 0.0663,
      "num_tokens": 624.0,
      "reward": 2.0061323642730713,
      "reward_std": 0.7225619554519653,
      "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": 1.5,
      "rewards/final_correctness_reward_func/std": 1.0,
      "rewards/question_recreation_reward_func/mean": 0.1031322255730629,
      "rewards/question_recreation_reward_func/std": 0.07210466265678406,
      "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": 0.40299999713897705,
      "rewards/xmlcount_reward_func/std": 0.035999998450279236,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 38.33333206176758,
      "completions/mean_terminated_length": 38.33333206176758,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 1.6666666666666665,
      "grad_norm": 11.897136688232422,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": -0.0037,
      "num_tokens": 1277.0,
      "reward": 1.6333832740783691,
      "reward_std": 0.0008207677165046334,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.25613313913345337,
      "rewards/question_recreation_reward_func/std": 0.23304274678230286,
      "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": 0.3772500157356262,
      "rewards/xmlcount_reward_func/std": 0.0505329966545105,
      "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": 26.0,
      "completions/max_terminated_length": 26.0,
      "completions/mean_length": 25.66666603088379,
      "completions/mean_terminated_length": 25.66666603088379,
      "completions/min_length": 25.0,
      "completions/min_terminated_length": 25.0,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.028560996055603027,
      "kl": 0.0006091607501730323,
      "learning_rate": 9.931806517013612e-07,
      "loss": 0.0,
      "num_tokens": 1891.0,
      "reward": 2.520005226135254,
      "reward_std": 0.0,
      "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": 2.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.10000516474246979,
      "rewards/question_recreation_reward_func/std": 0.058453768491744995,
      "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": 0.42000001668930054,
      "rewards/xmlcount_reward_func/std": 0.0011547028552740812,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 38.66666793823242,
      "completions/mean_terminated_length": 38.66666793823242,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 3.6666666666666665,
      "grad_norm": 10.408936500549316,
      "kl": 0.005873247544514015,
      "learning_rate": 9.729086208503173e-07,
      "loss": 0.0002,
      "num_tokens": 2545.0,
      "reward": 1.6425724029541016,
      "reward_std": 0.013538368977606297,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.2663222551345825,
      "rewards/question_recreation_reward_func/std": 0.2452709823846817,
      "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": 0.3762499988079071,
      "rewards/xmlcount_reward_func/std": 0.05167446658015251,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 42.33333206176758,
      "completions/mean_terminated_length": 42.33333206176758,
      "completions/min_length": 37.0,
      "completions/min_terminated_length": 37.0,
      "epoch": 4.666666666666667,
      "grad_norm": 15.399620056152344,
      "kl": 0.023564694682136178,
      "learning_rate": 9.397368756032444e-07,
      "loss": 0.0691,
      "num_tokens": 3209.0,
      "reward": 1.627357840538025,
      "reward_std": 0.012518457137048244,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.26235783100128174,
      "rewards/question_recreation_reward_func/std": 0.2709197998046875,
      "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": 0.36500000953674316,
      "rewards/xmlcount_reward_func/std": 0.04178517311811447,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 38.33333206176758,
      "completions/mean_terminated_length": 38.33333206176758,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 5.666666666666667,
      "grad_norm": 18.04197883605957,
      "kl": 0.026358011396951042,
      "learning_rate": 8.945702546981968e-07,
      "loss": -0.0029,
      "num_tokens": 3862.0,
      "reward": 1.6394498348236084,
      "reward_std": 0.006526038981974125,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.26219990849494934,
      "rewards/question_recreation_reward_func/std": 0.24019770324230194,
      "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": 0.3772500157356262,
      "rewards/xmlcount_reward_func/std": 0.0505329966545105,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 33.33333206176758,
      "completions/mean_terminated_length": 33.33333206176758,
      "completions/min_length": 25.0,
      "completions/min_terminated_length": 25.0,
      "epoch": 6.666666666666667,
      "grad_norm": 12.714645385742188,
      "kl": 0.008623177418485284,
      "learning_rate": 8.386407858128706e-07,
      "loss": -0.0318,
      "num_tokens": 4519.0,
      "reward": 1.463837742805481,
      "reward_std": 0.037738215178251266,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.09383778274059296,
      "rewards/question_recreation_reward_func/std": 0.06293194741010666,
      "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": 0.3700000047683716,
      "rewards/xmlcount_reward_func/std": 0.04478840157389641,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 32.33333206176758,
      "completions/mean_terminated_length": 32.33333206176758,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 7.666666666666667,
      "grad_norm": 0.1580970138311386,
      "kl": 0.0015887387562543154,
      "learning_rate": 7.734740790612136e-07,
      "loss": 0.0001,
      "num_tokens": 5173.0,
      "reward": 1.4761300086975098,
      "reward_std": 0.0,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.09962990880012512,
      "rewards/question_recreation_reward_func/std": 0.058887071907520294,
      "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": 0.3765000104904175,
      "rewards/xmlcount_reward_func/std": 0.05138418450951576,
      "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": 26.0,
      "completions/max_terminated_length": 26.0,
      "completions/mean_length": 25.33333396911621,
      "completions/mean_terminated_length": 25.33333396911621,
      "completions/min_length": 25.0,
      "completions/min_terminated_length": 25.0,
      "epoch": 8.666666666666666,
      "grad_norm": 0.3277643620967865,
      "kl": 0.004361201485153288,
      "learning_rate": 7.008477123264847e-07,
      "loss": 0.0002,
      "num_tokens": 5787.0,
      "reward": 2.4760446548461914,
      "reward_std": 0.0,
      "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": 2.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.056044623255729675,
      "rewards/question_recreation_reward_func/std": 0.015752185136079788,
      "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": 0.42000001668930054,
      "rewards/xmlcount_reward_func/std": 0.0011547028552740812,
      "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": 28.0,
      "completions/max_terminated_length": 28.0,
      "completions/mean_length": 26.66666603088379,
      "completions/mean_terminated_length": 26.66666603088379,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 9.666666666666666,
      "grad_norm": 0.5559372305870056,
      "kl": 0.41620227006205823,
      "learning_rate": 6.227427435703995e-07,
      "loss": 0.0166,
      "num_tokens": 6407.0,
      "reward": 1.510707139968872,
      "reward_std": 0.0,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.091707244515419,
      "rewards/question_recreation_reward_func/std": 0.06803537905216217,
      "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": 0.4190000295639038,
      "rewards/xmlcount_reward_func/std": 0.0023094057105481625,
      "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": 28.0,
      "completions/max_terminated_length": 28.0,
      "completions/mean_length": 27.33333396911621,
      "completions/mean_terminated_length": 27.33333396911621,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 10.666666666666666,
      "grad_norm": 37.45729064941406,
      "kl": 0.9323916549910791,
      "learning_rate": 5.412896727361662e-07,
      "loss": 0.0373,
      "num_tokens": 7027.0,
      "reward": 1.53065824508667,
      "reward_std": 0.0012277377536520362,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.11140821874141693,
      "rewards/question_recreation_reward_func/std": 0.06594393402338028,
      "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": 0.4192500114440918,
      "rewards/xmlcount_reward_func/std": 0.0020615539979189634,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 38.33333206176758,
      "completions/mean_terminated_length": 38.33333206176758,
      "completions/min_length": 25.0,
      "completions/min_terminated_length": 25.0,
      "epoch": 11.666666666666666,
      "grad_norm": 4.479156494140625,
      "kl": 0.017107000574469566,
      "learning_rate": 4.5871032726383385e-07,
      "loss": 0.0007,
      "num_tokens": 7679.0,
      "reward": 1.628129005432129,
      "reward_std": 0.0,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.25262898206710815,
      "rewards/question_recreation_reward_func/std": 0.26347851753234863,
      "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": 0.37549999356269836,
      "rewards/xmlcount_reward_func/std": 0.05022948235273361,
      "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": 26.0,
      "completions/max_terminated_length": 26.0,
      "completions/mean_length": 25.0,
      "completions/mean_terminated_length": 25.0,
      "completions/min_length": 24.0,
      "completions/min_terminated_length": 24.0,
      "epoch": 12.666666666666666,
      "grad_norm": 10.520891189575195,
      "kl": 0.08969685205374844,
      "learning_rate": 3.772572564296004e-07,
      "loss": -0.0036,
      "num_tokens": 8292.0,
      "reward": 2.4716968536376953,
      "reward_std": 0.006148550659418106,
      "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": 2.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.05094695836305618,
      "rewards/question_recreation_reward_func/std": 0.012908627279102802,
      "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": 0.4207499921321869,
      "rewards/xmlcount_reward_func/std": 0.0012583034113049507,
      "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": 26.0,
      "completions/max_terminated_length": 26.0,
      "completions/mean_length": 25.33333396911621,
      "completions/mean_terminated_length": 25.33333396911621,
      "completions/min_length": 25.0,
      "completions/min_terminated_length": 25.0,
      "epoch": 13.666666666666666,
      "grad_norm": 0.1571069359779358,
      "kl": 0.0024800227402010933,
      "learning_rate": 2.9915228767351535e-07,
      "loss": 0.0001,
      "num_tokens": 8906.0,
      "reward": 2.4760446548461914,
      "reward_std": 0.0,
      "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": 2.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.056044623255729675,
      "rewards/question_recreation_reward_func/std": 0.015752185136079788,
      "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": 0.42000001668930054,
      "rewards/xmlcount_reward_func/std": 0.0011547028552740812,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 38.0,
      "completions/mean_terminated_length": 38.0,
      "completions/min_length": 24.0,
      "completions/min_terminated_length": 24.0,
      "epoch": 14.666666666666666,
      "grad_norm": 10.73165512084961,
      "kl": 0.00937533297110349,
      "learning_rate": 2.2652592093878665e-07,
      "loss": 0.0074,
      "num_tokens": 9557.0,
      "reward": 1.6293513774871826,
      "reward_std": 0.001728863688185811,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.2538514733314514,
      "rewards/question_recreation_reward_func/std": 0.2620745301246643,
      "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": 0.37549999356269836,
      "rewards/xmlcount_reward_func/std": 0.05022948235273361,
      "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": 0.0,
      "completions/max_length": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 32.33333206176758,
      "completions/mean_terminated_length": 32.33333206176758,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 15.666666666666666,
      "grad_norm": 0.09817155450582504,
      "kl": 0.002157402690500021,
      "learning_rate": 1.6135921418712955e-07,
      "loss": 0.0001,
      "num_tokens": 10211.0,
      "reward": 1.4761300086975098,
      "reward_std": 0.0,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.09962990880012512,
      "rewards/question_recreation_reward_func/std": 0.058887071907520294,
      "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": 0.3765000104904175,
      "rewards/xmlcount_reward_func/std": 0.05138418450951576,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 40.0,
      "completions/mean_terminated_length": 40.0,
      "completions/min_length": 30.0,
      "completions/min_terminated_length": 30.0,
      "epoch": 16.666666666666668,
      "grad_norm": 11.78333854675293,
      "kl": 0.021301488392055035,
      "learning_rate": 1.0542974530180327e-07,
      "loss": 0.0327,
      "num_tokens": 10868.0,
      "reward": 1.6225154399871826,
      "reward_std": 0.007938506081700325,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.2535155117511749,
      "rewards/question_recreation_reward_func/std": 0.2625473141670227,
      "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": 0.36899998784065247,
      "rewards/xmlcount_reward_func/std": 0.04594925418496132,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 31.0,
      "completions/mean_terminated_length": 31.0,
      "completions/min_length": 23.0,
      "completions/min_terminated_length": 23.0,
      "epoch": 17.666666666666668,
      "grad_norm": 7.679316520690918,
      "kl": 0.056787674431689084,
      "learning_rate": 6.026312439675551e-08,
      "loss": 0.017,
      "num_tokens": 11518.0,
      "reward": 1.4841399192810059,
      "reward_std": 0.01688714697957039,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.10888981074094772,
      "rewards/question_recreation_reward_func/std": 0.05336005985736847,
      "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": 0.3752500116825104,
      "rewards/xmlcount_reward_func/std": 0.05022202059626579,
      "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": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 31.0,
      "completions/mean_terminated_length": 31.0,
      "completions/min_length": 24.0,
      "completions/min_terminated_length": 24.0,
      "epoch": 18.666666666666668,
      "grad_norm": 12.277668952941895,
      "kl": 0.08226607670076191,
      "learning_rate": 2.7091379149682682e-08,
      "loss": 0.0033,
      "num_tokens": 12168.0,
      "reward": 1.5170156955718994,
      "reward_std": 0.010529378429055214,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.14076562225818634,
      "rewards/question_recreation_reward_func/std": 0.0890238955616951,
      "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": 0.3762499988079071,
      "rewards/xmlcount_reward_func/std": 0.05109713599085808,
      "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": 0.0,
      "completions/max_length": 45.0,
      "completions/max_terminated_length": 45.0,
      "completions/mean_length": 38.66666793823242,
      "completions/mean_terminated_length": 38.66666793823242,
      "completions/min_length": 26.0,
      "completions/min_terminated_length": 26.0,
      "epoch": 19.666666666666668,
      "grad_norm": 0.03594563156366348,
      "kl": 0.002033954835496843,
      "learning_rate": 6.819348298638839e-09,
      "loss": 0.0001,
      "num_tokens": 12822.0,
      "reward": 1.6440644264221191,
      "reward_std": 0.0,
      "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": 1.0,
      "rewards/final_correctness_reward_func/std": 1.154700517654419,
      "rewards/question_recreation_reward_func/mean": 0.2675645053386688,
      "rewards/question_recreation_reward_func/std": 0.24623245000839233,
      "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": 0.3765000104904175,
      "rewards/xmlcount_reward_func/std": 0.05138418450951576,
      "step": 20
    },
    {
      "epoch": 19.666666666666668,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 0.010460580148537702,
      "train_runtime": 10202.6571,
      "train_samples_per_second": 0.008,
      "train_steps_per_second": 0.002
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 12822,
  "num_train_epochs": 20,
  "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
}