{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "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": 228.375,
      "epoch": 0.1,
      "grad_norm": 10.439698219299316,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": -0.0,
      "reward": 3.9033486885018647,
      "reward_std": 0.45634667729609646,
      "rewards/concensus_correctness_reward_func": 1.1147499978542328,
      "rewards/consensus_reward_func": 1.1875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.674067463260144,
      "rewards/soft_format_reward_func": 0.015625,
      "rewards/strict_format_reward_func": 0.234375,
      "rewards/xmlcount_reward_func": 0.6770312478765845,
      "step": 2
    },
    {
      "completion_length": 157.25,
      "epoch": 0.2,
      "grad_norm": 9.67512035369873,
      "kl": 0.27191234956262633,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0003,
      "reward": 6.323291674256325,
      "reward_std": 0.29528094810666516,
      "rewards/concensus_correctness_reward_func": 1.802937489002943,
      "rewards/consensus_reward_func": 1.9375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.8601979417726398,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.375,
      "rewards/xmlcount_reward_func": 1.22265625,
      "step": 4
    },
    {
      "completion_length": 183.59375,
      "epoch": 0.3,
      "grad_norm": 11.30721664428711,
      "kl": 0.7817973756464198,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0008,
      "reward": 6.385719895362854,
      "reward_std": 0.3340341460861964,
      "rewards/concensus_correctness_reward_func": 1.934749998152256,
      "rewards/consensus_reward_func": 1.875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.25,
      "rewards/question_recreation_reward_func": 0.8103449335321784,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.34375,
      "rewards/xmlcount_reward_func": 1.171875,
      "step": 6
    },
    {
      "completion_length": 150.96875,
      "epoch": 0.4,
      "grad_norm": 33.876792907714844,
      "kl": 2.9556224551051855,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.003,
      "reward": 6.251800790429115,
      "reward_std": 0.2621385550010018,
      "rewards/concensus_correctness_reward_func": 1.9013749873265624,
      "rewards/consensus_reward_func": 1.875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.8647695071995258,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.3125,
      "rewards/xmlcount_reward_func": 1.173156250268221,
      "step": 8
    },
    {
      "completion_length": 132.75,
      "epoch": 0.5,
      "grad_norm": 22.34031105041504,
      "kl": 1.533589490922168,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0015,
      "reward": 6.721142441034317,
      "reward_std": 0.37645364040508866,
      "rewards/concensus_correctness_reward_func": 2.0946249775588512,
      "rewards/consensus_reward_func": 1.9375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.25,
      "rewards/question_recreation_reward_func": 0.886954908259213,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.375,
      "rewards/xmlcount_reward_func": 1.177062500268221,
      "step": 10
    },
    {
      "completion_length": 147.625,
      "epoch": 0.6,
      "grad_norm": 43.14989471435547,
      "kl": 3.1210397497052327,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0031,
      "reward": 6.697916433215141,
      "reward_std": 0.357076933105418,
      "rewards/concensus_correctness_reward_func": 1.9758749939501286,
      "rewards/consensus_reward_func": 1.9375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.25,
      "rewards/question_recreation_reward_func": 0.8353225886821747,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.46875,
      "rewards/xmlcount_reward_func": 1.23046875,
      "step": 12
    },
    {
      "completion_length": 144.875,
      "epoch": 0.7,
      "grad_norm": 14.534745216369629,
      "kl": 32.27562204410788,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0323,
      "reward": 6.202968865633011,
      "reward_std": 0.3484449392417446,
      "rewards/concensus_correctness_reward_func": 1.815625011920929,
      "rewards/consensus_reward_func": 1.875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0625,
      "rewards/question_recreation_reward_func": 0.8974376134574413,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.390625,
      "rewards/xmlcount_reward_func": 1.1617812514305115,
      "step": 14
    },
    {
      "completion_length": 157.625,
      "epoch": 0.8,
      "grad_norm": 14.83914566040039,
      "kl": 4.667069488205016,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0047,
      "reward": 6.212416380643845,
      "reward_std": 0.5250872050528415,
      "rewards/concensus_correctness_reward_func": 1.806124985218048,
      "rewards/consensus_reward_func": 1.9375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.9003850966691971,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.34375,
      "rewards/xmlcount_reward_func": 1.0996562466025352,
      "step": 16
    },
    {
      "completion_length": 139.1875,
      "epoch": 0.9,
      "grad_norm": 17.672067642211914,
      "kl": 5.667459648102522,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0057,
      "reward": 6.300769180059433,
      "reward_std": 0.4447646054904908,
      "rewards/concensus_correctness_reward_func": 1.83356249704957,
      "rewards/consensus_reward_func": 1.8125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.25,
      "rewards/question_recreation_reward_func": 0.8667067214846611,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.359375,
      "rewards/xmlcount_reward_func": 1.178625002503395,
      "step": 18
    },
    {
      "completion_length": 138.3125,
      "epoch": 1.0,
      "grad_norm": 18.937057495117188,
      "kl": 7.192570003448054,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0072,
      "reward": 6.140490561723709,
      "reward_std": 0.5464329523674678,
      "rewards/concensus_correctness_reward_func": 1.854937493801117,
      "rewards/consensus_reward_func": 1.75,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.125,
      "rewards/question_recreation_reward_func": 0.8636780511587858,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.359375,
      "rewards/xmlcount_reward_func": 1.1875,
      "step": 20
    },
    {
      "epoch": 1.0,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 0.0058470307034440335,
      "train_runtime": 126.5029,
      "train_samples_per_second": 2.53,
      "train_steps_per_second": 0.158
    }
  ],
  "logging_steps": 2,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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
}