{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "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": 393.875,
      "epoch": 0.5714285714285714,
      "grad_norm": 14.467851638793945,
      "kl": 0.0036735038956976496,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.24942190013825893,
      "reward_std": 0.27551950167980976,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.14235940459184349,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.10706249950453639,
      "step": 2
    },
    {
      "completion_length": 499.3333333333333,
      "epoch": 1.0,
      "grad_norm": 2.420987844467163,
      "kl": 0.005199056991841644,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 0.5972021507720152,
      "reward_std": 0.5159138335535923,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.16666666666666666,
      "rewards/question_recreation_reward_func": 0.2659521692742904,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.16458333532015482,
      "step": 4
    },
    {
      "completion_length": 469.0625,
      "epoch": 1.5714285714285714,
      "grad_norm": 1.678683876991272,
      "kl": 0.0036473195723374374,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 0.2039470819872804,
      "reward_std": 0.41503745794761926,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.20882207114482298,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.12987499684095383,
      "step": 6
    },
    {
      "completion_length": 363.25,
      "epoch": 2.0,
      "grad_norm": 1.534175992012024,
      "kl": 0.0051646292946922285,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 1.6161055080592632,
      "reward_std": 3.3556958374877772,
      "rewards/concensus_correctness_reward_func": 1.6666666666666667,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.34002221499880153,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.3905833239356677,
      "step": 8
    },
    {
      "completion_length": 429.4375,
      "epoch": 2.571428571428571,
      "grad_norm": 4.225729942321777,
      "kl": 0.00387628185126232,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 0.604195196647197,
      "reward_std": 0.9071919927373528,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.5,
      "rewards/question_recreation_reward_func": 0.21207017498090863,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.10787498578429222,
      "step": 10
    },
    {
      "completion_length": 306.9166666666667,
      "epoch": 3.0,
      "grad_norm": 1.3296011686325073,
      "kl": 0.0014047033328097314,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 0.18260888879497847,
      "reward_std": 0.3342080342893799,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.14985888451337814,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.03275000055631002,
      "step": 12
    },
    {
      "completion_length": 509.0,
      "epoch": 3.571428571428571,
      "grad_norm": 1.647635579109192,
      "kl": 0.0013635017166961916,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 1.0174027022439986,
      "reward_std": 0.3876340059796348,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.5,
      "rewards/question_recreation_reward_func": 0.2895276902709156,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.2278749984689057,
      "step": 14
    },
    {
      "completion_length": 287.9166666666667,
      "epoch": 4.0,
      "grad_norm": 48.5843505859375,
      "kl": 0.0019373036241934944,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": -0.0794086183110873,
      "reward_std": 0.4189326539635658,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.10642471971611182,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.18583333492279053,
      "step": 16
    },
    {
      "completion_length": 402.75,
      "epoch": 4.571428571428571,
      "grad_norm": 16.052921295166016,
      "kl": 0.0023356471065199003,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 0.3587729576975107,
      "reward_std": 0.5114126596599817,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.19789796438999474,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.035875000059604645,
      "step": 18
    },
    {
      "completion_length": 331.25,
      "epoch": 5.0,
      "grad_norm": 3.976226806640625,
      "kl": 0.0043949687969870865,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 0.13239151674012342,
      "reward_std": 0.1440122425556183,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.06839151245852311,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.06399999931454659,
      "step": 20
    },
    {
      "epoch": 5.0,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 2.8325976700216417e-06,
      "train_runtime": 171.1424,
      "train_samples_per_second": 0.935,
      "train_steps_per_second": 0.117
    }
  ],
  "logging_steps": 2,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}