{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "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": 227.65625,
      "epoch": 0.5714285714285714,
      "grad_norm": 42.79217529296875,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 3.5435642525553703,
      "reward_std": 0.5061808098107576,
      "rewards/concensus_correctness_reward_func": 0.6306249964982271,
      "rewards/consensus_reward_func": 1.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.6840329738333821,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.203125,
      "rewards/xmlcount_reward_func": 0.9007812440395355,
      "step": 2
    },
    {
      "completion_length": 114.20833333333333,
      "epoch": 1.0,
      "grad_norm": 42.596649169921875,
      "kl": 0.17642021629338464,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0001,
      "reward": 6.350201050440471,
      "reward_std": 0.5205332829306523,
      "rewards/concensus_correctness_reward_func": 2.014166663090388,
      "rewards/consensus_reward_func": 1.5833333333333333,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.3333333333333333,
      "rewards/question_recreation_reward_func": 0.7852844769755999,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.4166666666666667,
      "rewards/xmlcount_reward_func": 1.2174166639645894,
      "step": 4
    },
    {
      "completion_length": 121.125,
      "epoch": 1.5714285714285714,
      "grad_norm": 54.49276351928711,
      "kl": 6.192452446790412,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0062,
      "reward": 5.543629631400108,
      "reward_std": 0.6034433040767908,
      "rewards/concensus_correctness_reward_func": 1.4573749955743551,
      "rewards/consensus_reward_func": 1.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.8040984030812979,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.421875,
      "rewards/xmlcount_reward_func": 1.2352812513709068,
      "step": 6
    },
    {
      "completion_length": 124.125,
      "epoch": 2.0,
      "grad_norm": 7102245.5,
      "kl": 126782.79426941897,
      "learning_rate": 3.867370395306068e-07,
      "loss": 95.0871,
      "reward": 6.043591817220052,
      "reward_std": 0.7726768641150557,
      "rewards/concensus_correctness_reward_func": 1.6195833186308544,
      "rewards/consensus_reward_func": 1.6666666666666667,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.25,
      "rewards/question_recreation_reward_func": 0.7885918033619722,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.4791666666666667,
      "rewards/xmlcount_reward_func": 1.2395833333333333,
      "step": 8
    },
    {
      "completion_length": 119.125,
      "epoch": 2.571428571428571,
      "grad_norm": 47.02878189086914,
      "kl": 14.295459616580047,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0143,
      "reward": 5.444987773895264,
      "reward_std": 1.0086145401000977,
      "rewards/concensus_correctness_reward_func": 1.4008749891072512,
      "rewards/consensus_reward_func": 1.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.1875,
      "rewards/question_recreation_reward_func": 0.7464564801193774,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.390625,
      "rewards/xmlcount_reward_func": 1.2195312529802322,
      "step": 10
    },
    {
      "completion_length": 113.58333333333333,
      "epoch": 3.0,
      "grad_norm": 38.122867584228516,
      "kl": 23412645888.984116,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 17559486.0,
      "reward": 6.110063831011455,
      "reward_std": 1.0958824685464303,
      "rewards/concensus_correctness_reward_func": 1.625583330790202,
      "rewards/consensus_reward_func": 1.5833333333333333,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.3333333333333333,
      "rewards/question_recreation_reward_func": 0.890730490287145,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.4375,
      "rewards/xmlcount_reward_func": 1.2395833333333333,
      "step": 12
    },
    {
      "completion_length": 116.15625,
      "epoch": 3.571428571428571,
      "grad_norm": 54.67125701904297,
      "kl": 2.0787589978426695,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0021,
      "reward": 6.153131395578384,
      "reward_std": 0.8237987434840761,
      "rewards/concensus_correctness_reward_func": 1.6433124989271164,
      "rewards/consensus_reward_func": 1.625,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.375,
      "rewards/question_recreation_reward_func": 0.8353501502424479,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.4375,
      "rewards/xmlcount_reward_func": 1.2369687482714653,
      "step": 14
    },
    {
      "completion_length": 126.16666666666667,
      "epoch": 4.0,
      "grad_norm": 21.411252975463867,
      "kl": 34.94235846462349,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0262,
      "reward": 5.952445189158122,
      "reward_std": 0.10796807702475537,
      "rewards/concensus_correctness_reward_func": 1.5421666478117306,
      "rewards/consensus_reward_func": 1.6666666666666667,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.16666666666666666,
      "rewards/question_recreation_reward_func": 0.8477784246206284,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.4791666666666667,
      "rewards/xmlcount_reward_func": 1.25,
      "step": 16
    },
    {
      "completion_length": 130.46875,
      "epoch": 4.571428571428571,
      "grad_norm": 97680.765625,
      "kl": 4103.804172766861,
      "learning_rate": 3.013156219837776e-08,
      "loss": 4.1038,
      "reward": 6.159731104969978,
      "reward_std": 0.27880998147884384,
      "rewards/concensus_correctness_reward_func": 1.8388749929144979,
      "rewards/consensus_reward_func": 1.5625,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.375,
      "rewards/question_recreation_reward_func": 0.7271060263738036,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.40625,
      "rewards/xmlcount_reward_func": 1.25,
      "step": 18
    },
    {
      "completion_length": 129.375,
      "epoch": 5.0,
      "grad_norm": 30.94512367248535,
      "kl": 35646709772.8303,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 26735034.0,
      "reward": 5.769796172777812,
      "reward_std": 0.2291471455731274,
      "rewards/concensus_correctness_reward_func": 1.51466666162014,
      "rewards/consensus_reward_func": 1.6666666666666667,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.08333333333333333,
      "rewards/question_recreation_reward_func": 0.8430045011142889,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.4375,
      "rewards/xmlcount_reward_func": 1.2246250013510387,
      "step": 20
    },
    {
      "epoch": 5.0,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 4429461.923979024,
      "train_runtime": 115.9765,
      "train_samples_per_second": 2.759,
      "train_steps_per_second": 0.172
    }
  ],
  "logging_steps": 2,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}