{
  "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": 393.0625,
      "epoch": 0.7272727272727273,
      "grad_norm": 8.715741157531738,
      "kl": 0.0007410245452774689,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 1.2119234781712294,
      "reward_std": 1.0487193930894136,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.4972672164440155,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": 0.3240312561392784,
      "step": 2
    },
    {
      "completion_length": 507.10714285714283,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.31629234552383423,
      "kl": 0.000937787736932348,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 1.166460822735514,
      "reward_std": 0.9958693981170654,
      "rewards/concensus_correctness_reward_func": 0.008928571428571428,
      "rewards/consensus_reward_func": 0.42857142857142855,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5971036766256604,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.13185714717422212,
      "step": 4
    },
    {
      "completion_length": 584.1071428571429,
      "epoch": 2.0,
      "grad_norm": 0.5126656889915466,
      "kl": 0.0006553483960617866,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 0.9678485904421125,
      "reward_std": 1.3451101311615534,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.42857142857142855,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5220271987574441,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.03571428571428571,
      "rewards/xmlcount_reward_func": -0.018464289605617523,
      "step": 6
    },
    {
      "completion_length": 567.78125,
      "epoch": 2.7272727272727275,
      "grad_norm": 0.3305479884147644,
      "kl": 0.0005583776874118485,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 1.132558684796095,
      "reward_std": 1.1547820568084717,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5389336231164634,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.09362500160932541,
      "step": 8
    },
    {
      "completion_length": 501.35714285714283,
      "epoch": 3.3636363636363638,
      "grad_norm": 7.692227363586426,
      "kl": 0.0010580410765084838,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 0.6616185222353254,
      "reward_std": 0.9527691198246819,
      "rewards/concensus_correctness_reward_func": -0.010499999991485051,
      "rewards/consensus_reward_func": 0.42857142857142855,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.41793995669909884,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.1743928444172655,
      "step": 10
    },
    {
      "completion_length": 462.2857142857143,
      "epoch": 4.0,
      "grad_norm": 0.2558252811431885,
      "kl": 0.0006222425228250879,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 0.7978931452546801,
      "reward_std": 1.0667394825390406,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.21428571428571427,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5569645762443542,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.02664285046713693,
      "step": 12
    },
    {
      "completion_length": 484.53125,
      "epoch": 4.7272727272727275,
      "grad_norm": 6.378965377807617,
      "kl": 0.0007858525641495362,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 1.1453090235590935,
      "reward_std": 1.1296067461371422,
      "rewards/concensus_correctness_reward_func": 0.0078125,
      "rewards/consensus_reward_func": 0.5625,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5136215183883905,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.06137500470504165,
      "step": 14
    },
    {
      "completion_length": 512.2857142857143,
      "epoch": 5.363636363636363,
      "grad_norm": 0.34200847148895264,
      "kl": 0.0006529915928175407,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": 0.781751641205379,
      "reward_std": 1.1218627776418413,
      "rewards/concensus_correctness_reward_func": 0.008928571428571428,
      "rewards/consensus_reward_func": 0.21428571428571427,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.555680194071361,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": -0.015000002724783761,
      "step": 16
    },
    {
      "completion_length": 476.17857142857144,
      "epoch": 6.0,
      "grad_norm": 0.33030903339385986,
      "kl": 0.0006651783977369112,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 0.8652351647615433,
      "reward_std": 0.8916095005614417,
      "rewards/concensus_correctness_reward_func": 0.0020714285118239267,
      "rewards/consensus_reward_func": 0.07142857142857142,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5398065458450999,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.03571428571428571,
      "rewards/xmlcount_reward_func": 0.2162142904209239,
      "step": 18
    },
    {
      "completion_length": 434.46875,
      "epoch": 6.7272727272727275,
      "grad_norm": 1.0812687873840332,
      "kl": 0.0014948861353332177,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 0.7459094524383545,
      "reward_std": 1.1404077112674713,
      "rewards/concensus_correctness_reward_func": 0.0078125,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.480628184042871,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": -0.13315625302493572,
      "step": 20
    },
    {
      "epoch": 6.7272727272727275,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 7.719418306351145e-07,
      "train_runtime": 1300.3846,
      "train_samples_per_second": 0.246,
      "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
}