{
  "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": 469.5,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.58329176902771,
      "kl": 0.000628454661637079,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 1.0317109785974026,
      "reward_std": 0.778160322457552,
      "rewards/concensus_correctness_reward_func": 0.010125000029802322,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.25,
      "rewards/question_recreation_reward_func": 0.5160547196865082,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": 0.23990623839199543,
      "step": 2
    },
    {
      "completion_length": 513.7142857142857,
      "epoch": 1.3636363636363638,
      "grad_norm": 29.813570022583008,
      "kl": 0.001547832091871117,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 1.2006304008620126,
      "reward_std": 1.2436990652765547,
      "rewards/concensus_correctness_reward_func": 0.05357142857142857,
      "rewards/consensus_reward_func": 0.2857142857142857,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5110232468162265,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.3324642777442932,
      "step": 4
    },
    {
      "completion_length": 580.5714285714286,
      "epoch": 2.0,
      "grad_norm": 5.044901371002197,
      "kl": 0.0005295541445125959,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 1.5192301060472215,
      "reward_std": 2.0026471508400783,
      "rewards/concensus_correctness_reward_func": 0.7142857142857143,
      "rewards/consensus_reward_func": 0.14285714285714285,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.21428571428571427,
      "rewards/question_recreation_reward_func": 0.49801578904901234,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.05021428795797484,
      "step": 6
    },
    {
      "completion_length": 629.5,
      "epoch": 2.7272727272727275,
      "grad_norm": 0.20881569385528564,
      "kl": 0.0005468809431476984,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 1.8400482200086117,
      "reward_std": 2.4333673045039177,
      "rewards/concensus_correctness_reward_func": 0.6796875,
      "rewards/consensus_reward_func": 0.3125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.6031420044600964,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.11971874348819256,
      "step": 8
    },
    {
      "completion_length": 466.9642857142857,
      "epoch": 3.3636363636363638,
      "grad_norm": 0.3577859699726105,
      "kl": 0.0006706916194941316,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 1.096388190984726,
      "reward_std": 0.8796266477022853,
      "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": 0.4578524577830519,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.424249991774559,
      "step": 10
    },
    {
      "completion_length": 556.1785714285714,
      "epoch": 4.0,
      "grad_norm": 0.40372711420059204,
      "kl": 0.0006262019160203636,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 2.1709934047290256,
      "reward_std": 3.0565604141780307,
      "rewards/concensus_correctness_reward_func": 0.7678571428571429,
      "rewards/consensus_reward_func": 0.42857142857142855,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.21428571428571427,
      "rewards/question_recreation_reward_func": 0.5717433776174273,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.18853571744901793,
      "step": 12
    },
    {
      "completion_length": 473.875,
      "epoch": 4.7272727272727275,
      "grad_norm": 2.351263999938965,
      "kl": 0.0006538540837937035,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 1.398249976336956,
      "reward_std": 1.1338147409260273,
      "rewards/concensus_correctness_reward_func": 0.09375,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5107187051326036,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": 0.40315625444054604,
      "step": 14
    },
    {
      "completion_length": 611.0714285714286,
      "epoch": 5.363636363636363,
      "grad_norm": 0.3027333617210388,
      "kl": 0.0005950834435809936,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": 1.858400855745588,
      "reward_std": 2.591935762337276,
      "rewards/concensus_correctness_reward_func": 0.7678571428571429,
      "rewards/consensus_reward_func": 0.35714285714285715,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.14285714285714285,
      "rewards/question_recreation_reward_func": 0.4938651578766959,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.09667855714048658,
      "step": 16
    },
    {
      "completion_length": 519.8571428571429,
      "epoch": 6.0,
      "grad_norm": 0.403927206993103,
      "kl": 0.0006464655161835253,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 1.0676009271826064,
      "reward_std": 0.7614234515598842,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.14285714285714285,
      "rewards/question_recreation_reward_func": 0.5988509144101825,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.3080357126891613,
      "step": 18
    },
    {
      "completion_length": 541.53125,
      "epoch": 6.7272727272727275,
      "grad_norm": 0.547307550907135,
      "kl": 0.0006191184584167786,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 1.588839590549469,
      "reward_std": 2.0721157900989056,
      "rewards/concensus_correctness_reward_func": 0.671875,
      "rewards/consensus_reward_func": 0.125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.25,
      "rewards/question_recreation_reward_func": 0.6057458873838186,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": -0.0637812539935112,
      "step": 20
    },
    {
      "epoch": 6.7272727272727275,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 6.379948018775394e-07,
      "train_runtime": 1515.4809,
      "train_samples_per_second": 0.211,
      "train_steps_per_second": 0.013
    }
  ],
  "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
}