{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 19.571428571428573,
  "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": 250.98684210526315,
      "epoch": 1.5714285714285714,
      "grad_norm": 35.16233825683594,
      "kl": 27.098115991604956,
      "learning_rate": 4.965903258506806e-07,
      "loss": 0.0322,
      "reward": 2.5767549056755867,
      "reward_std": 1.3109975865012722,
      "rewards/concensus_correctness_reward_func": 1.0325000160618831,
      "rewards/consensus_reward_func": 0.6578947368421053,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.02631578947368421,
      "rewards/question_recreation_reward_func": 0.25305746799628986,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.05263157894736842,
      "rewards/xmlcount_reward_func": 0.5543552668471086,
      "step": 2
    },
    {
      "completion_length": 139.48863636363637,
      "epoch": 3.571428571428571,
      "grad_norm": 44.14933395385742,
      "kl": 7.791636250235817,
      "learning_rate": 4.698684378016222e-07,
      "loss": 0.0107,
      "reward": 2.973739290779287,
      "reward_std": 1.2925651385025545,
      "rewards/concensus_correctness_reward_func": 0.8166136467321352,
      "rewards/consensus_reward_func": 0.8863636363636364,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.022727272727272728,
      "rewards/question_recreation_reward_func": 0.2649438115344806,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.07386363636363637,
      "rewards/xmlcount_reward_func": 0.9092272858728062,
      "step": 4
    },
    {
      "completion_length": 133.11363636363637,
      "epoch": 5.571428571428571,
      "grad_norm": 48.02810287475586,
      "kl": 11.512206840244206,
      "learning_rate": 4.193203929064353e-07,
      "loss": 0.0158,
      "reward": 2.8814289461482656,
      "reward_std": 1.2390722137960521,
      "rewards/concensus_correctness_reward_func": 0.748363643545996,
      "rewards/consensus_reward_func": 0.7272727272727273,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.06818181818181818,
      "rewards/question_recreation_reward_func": 0.29970166425813327,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0625,
      "rewards/xmlcount_reward_func": 0.9754091013561595,
      "step": 6
    },
    {
      "completion_length": 141.04545454545453,
      "epoch": 7.571428571428571,
      "grad_norm": 38.969947814941406,
      "kl": 12.051131608811291,
      "learning_rate": 3.5042385616324236e-07,
      "loss": 0.0166,
      "reward": 3.381654983217066,
      "reward_std": 1.4018883820284496,
      "rewards/concensus_correctness_reward_func": 0.8803863657469099,
      "rewards/consensus_reward_func": 1.1136363636363635,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.06818181818181818,
      "rewards/question_recreation_reward_func": 0.26520041410218587,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.09090909090909091,
      "rewards/xmlcount_reward_func": 0.9633409245447679,
      "step": 8
    },
    {
      "completion_length": 149.89772727272728,
      "epoch": 9.571428571428571,
      "grad_norm": 24.2377872467041,
      "kl": 14.045930672775615,
      "learning_rate": 2.706448363680831e-07,
      "loss": 0.0193,
      "reward": 3.2756326740438286,
      "reward_std": 1.2956370193172584,
      "rewards/concensus_correctness_reward_func": 0.9670681904324077,
      "rewards/consensus_reward_func": 0.9545454545454546,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.1590909090909091,
      "rewards/question_recreation_reward_func": 0.2421099069443616,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0625,
      "rewards/xmlcount_reward_func": 0.8903181905096228,
      "step": 10
    },
    {
      "completion_length": 141.95454545454547,
      "epoch": 11.571428571428571,
      "grad_norm": 29.485441207885742,
      "kl": 32.70376639745452,
      "learning_rate": 1.886286282148002e-07,
      "loss": 0.045,
      "reward": 3.226967532526363,
      "reward_std": 1.3467150957069614,
      "rewards/concensus_correctness_reward_func": 0.9230227328159593,
      "rewards/consensus_reward_func": 1.0227272727272727,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.06818181818181818,
      "rewards/question_recreation_reward_func": 0.3108766495504163,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.056818181818181816,
      "rewards/xmlcount_reward_func": 0.8453409211202101,
      "step": 12
    },
    {
      "completion_length": 160.9659090909091,
      "epoch": 13.571428571428571,
      "grad_norm": 248509.75,
      "kl": 10224.36121920022,
      "learning_rate": 1.1326296046939333e-07,
      "loss": 14.0585,
      "reward": 3.0833815173669294,
      "reward_std": 1.6619971624829553,
      "rewards/concensus_correctness_reward_func": 0.9169318371198394,
      "rewards/consensus_reward_func": 0.8863636363636364,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.045454545454545456,
      "rewards/question_recreation_reward_func": 0.2725861096246676,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.05113636363636364,
      "rewards/xmlcount_reward_func": 0.9109090945937417,
      "step": 14
    },
    {
      "completion_length": 136.02272727272728,
      "epoch": 15.571428571428571,
      "grad_norm": 31.18145179748535,
      "kl": 22.365940256552264,
      "learning_rate": 5.271487265090163e-08,
      "loss": 0.0308,
      "reward": 3.1623240330002527,
      "reward_std": 1.3216449283063412,
      "rewards/concensus_correctness_reward_func": 0.9653409263965759,
      "rewards/consensus_reward_func": 0.8409090909090909,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.09090909090909091,
      "rewards/question_recreation_reward_func": 0.23630127632482487,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.06818181818181818,
      "rewards/xmlcount_reward_func": 0.9606818367134441,
      "step": 16
    },
    {
      "completion_length": 146.86363636363637,
      "epoch": 17.571428571428573,
      "grad_norm": 39.81138229370117,
      "kl": 18.526999665932223,
      "learning_rate": 1.3545689574841341e-08,
      "loss": 0.0255,
      "reward": 3.284146102991971,
      "reward_std": 1.468788014894182,
      "rewards/concensus_correctness_reward_func": 1.177477299828421,
      "rewards/consensus_reward_func": 0.75,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.045454545454545456,
      "rewards/question_recreation_reward_func": 0.22682786627079954,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.09659090909090909,
      "rewards/xmlcount_reward_func": 0.9877954667264764,
      "step": 18
    },
    {
      "completion_length": 141.5909090909091,
      "epoch": 19.571428571428573,
      "grad_norm": 36.41402816772461,
      "kl": 89.27689167044379,
      "learning_rate": 0.0,
      "loss": 0.1228,
      "reward": 3.6040925925428215,
      "reward_std": 1.723052983934229,
      "rewards/concensus_correctness_reward_func": 1.141272744671865,
      "rewards/consensus_reward_func": 1.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.1590909090909091,
      "rewards/question_recreation_reward_func": 0.28827438029375946,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.06818181818181818,
      "rewards/xmlcount_reward_func": 0.9472727450457487,
      "step": 20
    },
    {
      "epoch": 19.571428571428573,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 1.4377086507156491,
      "train_runtime": 480.0482,
      "train_samples_per_second": 1.333,
      "train_steps_per_second": 0.042
    }
  ],
  "logging_steps": 2,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "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": 4,
  "trial_name": null,
  "trial_params": null
}