{
  "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": 152.6875,
      "epoch": 0.5714285714285714,
      "grad_norm": 9.704741898985958e-09,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 7.741498485207558,
      "reward_std": 0.0,
      "rewards/concensus_correctness_reward_func": 2.0381250008940697,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.125,
      "rewards/question_recreation_reward_func": 0.9377485089353286,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.46875,
      "rewards/xmlcount_reward_func": 1.171875,
      "step": 2
    },
    {
      "completion_length": 166.125,
      "epoch": 1.0,
      "grad_norm": 0.00010551138257142156,
      "kl": 6.533205315906176e-07,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 8.665034532546997,
      "reward_std": 0.0,
      "rewards/concensus_correctness_reward_func": 2.248999993006388,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.6666666666666667,
      "rewards/question_recreation_reward_func": 0.9993677288293839,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.5,
      "rewards/xmlcount_reward_func": 1.25,
      "step": 4
    },
    {
      "completion_length": 152.0625,
      "epoch": 1.5714285714285714,
      "grad_norm": 1.6149582862854004,
      "kl": 0.004159657626122026,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 7.852498576045036,
      "reward_std": 0.022097086533904076,
      "rewards/concensus_correctness_reward_func": 2.164750002324581,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.125,
      "rewards/question_recreation_reward_func": 0.9377485089353286,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.453125,
      "rewards/xmlcount_reward_func": 1.171875,
      "step": 6
    },
    {
      "completion_length": 173.0,
      "epoch": 2.0,
      "grad_norm": 1.253675103187561,
      "kl": 0.009927107773364696,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 8.412867744763693,
      "reward_std": 0.3535533547401428,
      "rewards/concensus_correctness_reward_func": 2.0801666577657065,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.5833333333333333,
      "rewards/question_recreation_reward_func": 0.9993677288293839,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.5,
      "rewards/xmlcount_reward_func": 1.25,
      "step": 8
    },
    {
      "completion_length": 166.59375,
      "epoch": 2.571428571428571,
      "grad_norm": 0.003849708242341876,
      "kl": 0.010342858462536242,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 7.742899313569069,
      "reward_std": 0.0,
      "rewards/concensus_correctness_reward_func": 2.1649999991059303,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.0,
      "rewards/question_recreation_reward_func": 0.9372743055573665,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.46875,
      "rewards/xmlcount_reward_func": 1.171875,
      "step": 10
    },
    {
      "completion_length": 160.79166666666666,
      "epoch": 3.0,
      "grad_norm": 1.604172945022583,
      "kl": 0.009363556828854295,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 8.413166761398315,
      "reward_std": 0.3535534143447876,
      "rewards/concensus_correctness_reward_func": 2.0798333287239075,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.5833333333333333,
      "rewards/question_recreation_reward_func": 1.0,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.5,
      "rewards/xmlcount_reward_func": 1.25,
      "step": 12
    },
    {
      "completion_length": 156.125,
      "epoch": 3.571428571428571,
      "grad_norm": 1.433937668800354,
      "kl": 0.013254275647341274,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 8.117686077952385,
      "reward_std": 0.176865161061869,
      "rewards/concensus_correctness_reward_func": 2.1643749997019768,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.375,
      "rewards/question_recreation_reward_func": 0.9377485089353286,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.46875,
      "rewards/xmlcount_reward_func": 1.171812504529953,
      "step": 14
    },
    {
      "completion_length": 172.5,
      "epoch": 4.0,
      "grad_norm": 1.5800690650939941,
      "kl": 0.007285505174271141,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": 7.917034427324931,
      "reward_std": 0.34836794932683307,
      "rewards/concensus_correctness_reward_func": 2.000999997059504,
      "rewards/consensus_reward_func": 1.9166666666666667,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.25,
      "rewards/question_recreation_reward_func": 0.9993677288293839,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.5,
      "rewards/xmlcount_reward_func": 1.25,
      "step": 16
    },
    {
      "completion_length": 150.96875,
      "epoch": 4.571428571428571,
      "grad_norm": 0.009897509589791298,
      "kl": 0.007831997059838613,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 8.117623582482338,
      "reward_std": 0.1767767146229744,
      "rewards/concensus_correctness_reward_func": 2.1642499938607216,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.375,
      "rewards/question_recreation_reward_func": 0.9377485089353286,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.46875,
      "rewards/xmlcount_reward_func": 1.171875,
      "step": 18
    },
    {
      "completion_length": 177.08333333333334,
      "epoch": 5.0,
      "grad_norm": 1.1512504816055298,
      "kl": 0.00556948969218259,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 8.246867775917053,
      "reward_std": 0.11785115798314412,
      "rewards/concensus_correctness_reward_func": 2.0808333357175193,
      "rewards/consensus_reward_func": 2.0,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.4166666666666667,
      "rewards/question_recreation_reward_func": 0.9993677288293839,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.5,
      "rewards/xmlcount_reward_func": 1.25,
      "step": 20
    },
    {
      "epoch": 5.0,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 5.9759746136645034e-06,
      "train_runtime": 2064.6435,
      "train_samples_per_second": 0.155,
      "train_steps_per_second": 0.01
    }
  ],
  "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
}