{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.7272727272727275,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 506.6875,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.4768719971179962,
      "kl": 0.0006958687226870097,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 1.8010706398636103,
      "reward_std": 2.413410998880863,
      "rewards/concensus_correctness_reward_func": 1.2969374991953373,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0625,
      "rewards/question_recreation_reward_func": 0.5273518972098827,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.046875,
      "rewards/xmlcount_reward_func": -0.13259375002235174,
      "step": 2
    },
    {
      "completion_length": 486.7857142857143,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.33231866359710693,
      "kl": 0.0006329953279678843,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 0.918908280985696,
      "reward_std": 0.9659519238131387,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.2857142857142857,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5362654126116208,
      "rewards/soft_format_reward_func": 0.017857142857142856,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.07907143448080335,
      "step": 4
    },
    {
      "completion_length": 416.42857142857144,
      "epoch": 2.0,
      "grad_norm": 0.49434342980384827,
      "kl": 0.0008667865518613585,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 2.337849646806717,
      "reward_std": 2.588186740875244,
      "rewards/concensus_correctness_reward_func": 1.429999999968069,
      "rewards/consensus_reward_func": 0.21428571428571427,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.14285714285714285,
      "rewards/question_recreation_reward_func": 0.40942103415727615,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.14128571322986058,
      "step": 6
    },
    {
      "completion_length": 475.03125,
      "epoch": 2.7272727272727275,
      "grad_norm": 0.3978624641895294,
      "kl": 0.0007225602130347397,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 0.712943073362112,
      "reward_std": 0.844877365976572,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.3125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5247243288904428,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": -0.13990625087171793,
      "step": 8
    },
    {
      "completion_length": 457.17857142857144,
      "epoch": 3.3636363636363638,
      "grad_norm": 0.48263752460479736,
      "kl": 0.0008737414692794638,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 1.424566850066185,
      "reward_std": 1.2323028487818581,
      "rewards/concensus_correctness_reward_func": 0.1630000046321324,
      "rewards/consensus_reward_func": 0.42857142857142855,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.14285714285714285,
      "rewards/question_recreation_reward_func": 0.5036740648959365,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.1864642830831664,
      "step": 10
    },
    {
      "completion_length": 404.17857142857144,
      "epoch": 4.0,
      "grad_norm": 1.2086650133132935,
      "kl": 0.0008574270676555378,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 1.2912089100905828,
      "reward_std": 1.2101118607180459,
      "rewards/concensus_correctness_reward_func": 0.0014285713966403688,
      "rewards/consensus_reward_func": 0.35714285714285715,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.21428571428571427,
      "rewards/question_recreation_reward_func": 0.5251731936420713,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.17532142837132728,
      "step": 12
    },
    {
      "completion_length": 478.6875,
      "epoch": 4.7272727272727275,
      "grad_norm": 0.45308005809783936,
      "kl": 0.0007028918153082486,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 2.8731972258538008,
      "reward_std": 2.0956926345825195,
      "rewards/concensus_correctness_reward_func": 1.875,
      "rewards/consensus_reward_func": 0.1875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.5900409906171262,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": 0.0800312552601099,
      "step": 14
    },
    {
      "completion_length": 570.0,
      "epoch": 5.363636363636363,
      "grad_norm": 0.32726794481277466,
      "kl": 0.0006596180735089417,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": 1.9665765294006892,
      "reward_std": 2.4709178635052274,
      "rewards/concensus_correctness_reward_func": 1.429999999968069,
      "rewards/consensus_reward_func": 0.07142857142857142,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.14285714285714285,
      "rewards/question_recreation_reward_func": 0.4725050444581679,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.15021428252969468,
      "step": 16
    },
    {
      "completion_length": 474.10714285714283,
      "epoch": 6.0,
      "grad_norm": 0.3391857445240021,
      "kl": 0.0007322329412480551,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 0.8521491778748376,
      "reward_std": 0.8532949728625161,
      "rewards/concensus_correctness_reward_func": 0.0689285716840199,
      "rewards/consensus_reward_func": 0.14285714285714285,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5392563108886991,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.10110714712313243,
      "step": 18
    },
    {
      "completion_length": 452.90625,
      "epoch": 6.7272727272727275,
      "grad_norm": 23.497669219970703,
      "kl": 0.002060862439975608,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 0.922668170183897,
      "reward_std": 1.0907279998064041,
      "rewards/concensus_correctness_reward_func": 0.07512500137090683,
      "rewards/consensus_reward_func": 0.125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0625,
      "rewards/question_recreation_reward_func": 0.5273868888616562,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.046875,
      "rewards/xmlcount_reward_func": 0.08578124456107616,
      "step": 20
    },
    {
      "epoch": 6.7272727272727275,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 8.18494260101943e-07,
      "train_runtime": 1354.9062,
      "train_samples_per_second": 0.236,
      "train_steps_per_second": 0.015
    }
  ],
  "logging_steps": 2,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}