{
  "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": 495.34375,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.3851245939731598,
      "kl": 0.0015201627466012724,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 1.23531174659729,
      "reward_std": 1.1557490825653076,
      "rewards/concensus_correctness_reward_func": 0.0008750000270083547,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0625,
      "rewards/question_recreation_reward_func": 0.6263742372393608,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.1705624982714653,
      "step": 2
    },
    {
      "completion_length": 362.7142857142857,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.46032318472862244,
      "kl": 0.0023172974719532897,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 1.392972162791661,
      "reward_std": 1.4584640605109078,
      "rewards/concensus_correctness_reward_func": 0.14257142799241201,
      "rewards/consensus_reward_func": 0.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.3609007064785276,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.3895000049046108,
      "step": 4
    },
    {
      "completion_length": 377.7857142857143,
      "epoch": 2.0,
      "grad_norm": 3.0854783058166504,
      "kl": 0.00172044662758708,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 0.6036367884704045,
      "reward_std": 0.5871127205235618,
      "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": 0.5374582058617047,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.0661785677075386,
      "step": 6
    },
    {
      "completion_length": 432.21875,
      "epoch": 2.7272727272727275,
      "grad_norm": 0.4442577660083771,
      "kl": 0.0016866580190253444,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 1.3614656776189804,
      "reward_std": 1.080071672797203,
      "rewards/concensus_correctness_reward_func": 0.03099999949336052,
      "rewards/consensus_reward_func": 0.4375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5839969431981444,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.30896875262260437,
      "step": 8
    },
    {
      "completion_length": 349.7142857142857,
      "epoch": 3.3636363636363638,
      "grad_norm": 0.5599241852760315,
      "kl": 0.0020511089491524865,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 1.660039212022509,
      "reward_std": 1.1306958198547363,
      "rewards/concensus_correctness_reward_func": 0.30899999822889057,
      "rewards/consensus_reward_func": 0.5714285714285714,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.34428915807179045,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.03571428571428571,
      "rewards/xmlcount_reward_func": 0.39960715600422453,
      "step": 10
    },
    {
      "completion_length": 495.92857142857144,
      "epoch": 4.0,
      "grad_norm": 0.38454505801200867,
      "kl": 0.0019243384096106248,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 1.0013103996004378,
      "reward_std": 0.9705695403473718,
      "rewards/concensus_correctness_reward_func": 0.12135714292526245,
      "rewards/consensus_reward_func": 0.2857142857142857,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5770603716373444,
      "rewards/soft_format_reward_func": 0.017857142857142856,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.000678571207182748,
      "step": 12
    },
    {
      "completion_length": 403.53125,
      "epoch": 4.7272727272727275,
      "grad_norm": 0.4447067677974701,
      "kl": 0.0018273314417456277,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 1.136631891131401,
      "reward_std": 0.8802363304421306,
      "rewards/concensus_correctness_reward_func": 0.07368750125169754,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.37563188280910254,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.3123125210404396,
      "step": 14
    },
    {
      "completion_length": 439.35714285714283,
      "epoch": 5.363636363636363,
      "grad_norm": 0.4888899028301239,
      "kl": 0.0017850026072535133,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": 0.7386083092008319,
      "reward_std": 0.9949612553630557,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.07142857142857142,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.4658940147076334,
      "rewards/soft_format_reward_func": 0.017857142857142856,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.16557144692965917,
      "step": 16
    },
    {
      "completion_length": 390.10714285714283,
      "epoch": 6.0,
      "grad_norm": 0.4372193217277527,
      "kl": 0.0022854185164240854,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 1.2345900194985526,
      "reward_std": 1.5980539066450936,
      "rewards/concensus_correctness_reward_func": 0.11921428782599312,
      "rewards/consensus_reward_func": 0.42857142857142855,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.49841142126492094,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.18839287012815475,
      "step": 18
    },
    {
      "completion_length": 411.34375,
      "epoch": 6.7272727272727275,
      "grad_norm": 0.7081403136253357,
      "kl": 0.0019678706958075054,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 0.8901518955826759,
      "reward_std": 0.9801672622561455,
      "rewards/concensus_correctness_reward_func": 0.04268750175833702,
      "rewards/consensus_reward_func": 0.125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.4092768579721451,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": 0.29756249696947634,
      "step": 20
    },
    {
      "epoch": 6.7272727272727275,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 1.760407212714199e-06,
      "train_runtime": 560.1249,
      "train_samples_per_second": 0.571,
      "train_steps_per_second": 0.036
    }
  ],
  "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
}