{
  "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": 501.3125,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.41821882128715515,
      "kl": 0.0021961630882287864,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.997173797339201,
      "reward_std": 0.872414430603385,
      "rewards/concensus_correctness_reward_func": 0.01656249922234565,
      "rewards/consensus_reward_func": 0.1875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.671673808246851,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.12143751420080662,
      "step": 2
    },
    {
      "completion_length": 593.7857142857143,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.3008968234062195,
      "kl": 0.0015924011240713298,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 1.1429157044206346,
      "reward_std": 1.1326338393347604,
      "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": 0.570058537381036,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.14428571718079702,
      "step": 4
    },
    {
      "completion_length": 690.1785714285714,
      "epoch": 2.0,
      "grad_norm": 0.4040604829788208,
      "kl": 0.002621982429575707,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 1.3513478807040624,
      "reward_std": 1.1190883389541082,
      "rewards/concensus_correctness_reward_func": 0.034285713519368856,
      "rewards/consensus_reward_func": 0.2857142857142857,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.6456336123602731,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.3857142946549824,
      "step": 6
    },
    {
      "completion_length": 558.65625,
      "epoch": 2.7272727272727275,
      "grad_norm": 0.40765684843063354,
      "kl": 0.0018274225913046394,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 1.2382279224693775,
      "reward_std": 1.1536910682916641,
      "rewards/concensus_correctness_reward_func": 0.05718749953666702,
      "rewards/consensus_reward_func": 0.3125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.6502904426306486,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.21824999805539846,
      "step": 8
    },
    {
      "completion_length": 602.1071428571429,
      "epoch": 3.3636363636363638,
      "grad_norm": 0.38519588112831116,
      "kl": 0.0021380752690934707,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 1.461499001298632,
      "reward_std": 1.3494994640350342,
      "rewards/concensus_correctness_reward_func": 0.05964285561016628,
      "rewards/consensus_reward_func": 0.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.7344633000237601,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.14953572409493582,
      "step": 10
    },
    {
      "completion_length": 613.6785714285714,
      "epoch": 4.0,
      "grad_norm": 0.33382368087768555,
      "kl": 0.0012572459776752762,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 0.7477920481136867,
      "reward_std": 0.9433660166604179,
      "rewards/concensus_correctness_reward_func": 0.03242857115609305,
      "rewards/consensus_reward_func": 0.21428571428571427,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.460077733865806,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.04099997665200915,
      "step": 12
    },
    {
      "completion_length": 620.59375,
      "epoch": 4.7272727272727275,
      "grad_norm": 0.3225250244140625,
      "kl": 0.0021717731578974053,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 1.6406545080244541,
      "reward_std": 1.2635152488946915,
      "rewards/concensus_correctness_reward_func": 0.05056249909102917,
      "rewards/consensus_reward_func": 0.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.6221232824027538,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.46796876192092896,
      "step": 14
    },
    {
      "completion_length": 653.9642857142857,
      "epoch": 5.363636363636363,
      "grad_norm": 0.3691648542881012,
      "kl": 0.0015798391292004713,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": 1.3272171579301357,
      "reward_std": 1.0728605687618256,
      "rewards/concensus_correctness_reward_func": 0.10621428489685059,
      "rewards/consensus_reward_func": 0.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5657171458005905,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.15528571712119238,
      "step": 16
    },
    {
      "completion_length": 550.5714285714286,
      "epoch": 6.0,
      "grad_norm": 0.5164164900779724,
      "kl": 0.002296847482544503,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 1.4456927904060908,
      "reward_std": 1.255401929042169,
      "rewards/concensus_correctness_reward_func": 0.05121428419702819,
      "rewards/consensus_reward_func": 0.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.6957642180579049,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.1808571464249066,
      "step": 18
    },
    {
      "completion_length": 546.3125,
      "epoch": 6.7272727272727275,
      "grad_norm": 3.3458011150360107,
      "kl": 0.0020920391543768346,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 1.202021263539791,
      "reward_std": 1.3566488791257143,
      "rewards/concensus_correctness_reward_func": 0.04437499865889549,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.6228962391614914,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.03125,
      "rewards/xmlcount_reward_func": 0.12850000709295273,
      "step": 20
    },
    {
      "epoch": 6.7272727272727275,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 1.8512214751353896e-06,
      "train_runtime": 719.4431,
      "train_samples_per_second": 0.445,
      "train_steps_per_second": 0.028
    }
  ],
  "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
}