{
  "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": 522.875,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.36991098523139954,
      "kl": 0.0019559332286007702,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.7413240652531385,
      "reward_std": 1.150701311416924,
      "rewards/concensus_correctness_reward_func": 0.026125000789761543,
      "rewards/consensus_reward_func": 0.125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5951052904129028,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": -0.020531244575977325,
      "step": 2
    },
    {
      "completion_length": 471.0,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.4256592094898224,
      "kl": 0.002384057900469218,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0,
      "reward": 0.7693858827863421,
      "reward_std": 1.2164284203733717,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.35714285714285715,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.07142857142857142,
      "rewards/question_recreation_reward_func": 0.623278768999236,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": -0.3003214235816683,
      "step": 4
    },
    {
      "completion_length": 499.07142857142856,
      "epoch": 2.0,
      "grad_norm": 0.3912314176559448,
      "kl": 0.0020282766449132134,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.0,
      "reward": 1.2352780273982458,
      "reward_std": 1.0975747534206934,
      "rewards/concensus_correctness_reward_func": 0.0538571434361594,
      "rewards/consensus_reward_func": 0.35714285714285715,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.6258851566484996,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.03571428571428571,
      "rewards/xmlcount_reward_func": 0.16267855252538407,
      "step": 6
    },
    {
      "completion_length": 460.96875,
      "epoch": 2.7272727272727275,
      "grad_norm": 0.4590853452682495,
      "kl": 0.002819538931362331,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0,
      "reward": 1.2401280123740435,
      "reward_std": 1.189345095306635,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0625,
      "rewards/question_recreation_reward_func": 0.6354092583060265,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.046875,
      "rewards/xmlcount_reward_func": 0.12034375872462988,
      "step": 8
    },
    {
      "completion_length": 474.32142857142856,
      "epoch": 3.3636363636363638,
      "grad_norm": 0.5141214728355408,
      "kl": 0.002193358710168728,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.0,
      "reward": 1.1957727095910482,
      "reward_std": 1.3519693613052368,
      "rewards/concensus_correctness_reward_func": 0.06557142947401319,
      "rewards/consensus_reward_func": 0.42857142857142855,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5845941177436283,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.09917857976896423,
      "step": 10
    },
    {
      "completion_length": 461.0357142857143,
      "epoch": 4.0,
      "grad_norm": 0.3392316997051239,
      "kl": 0.002619804282273565,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.0,
      "reward": 0.9551736222846168,
      "reward_std": 0.8036346776144845,
      "rewards/concensus_correctness_reward_func": 0.0,
      "rewards/consensus_reward_func": 0.14285714285714285,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5953164739268166,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.19914287328720093,
      "step": 12
    },
    {
      "completion_length": 450.0625,
      "epoch": 4.7272727272727275,
      "grad_norm": 0.6728530526161194,
      "kl": 0.0024441872228635475,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.0,
      "reward": 0.9758850112557411,
      "reward_std": 1.0882981829345226,
      "rewards/concensus_correctness_reward_func": 0.05737500078976154,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.5361974686384201,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.03125,
      "rewards/xmlcount_reward_func": -0.023937483318150043,
      "step": 14
    },
    {
      "completion_length": 463.7857142857143,
      "epoch": 5.363636363636363,
      "grad_norm": 0.31825771927833557,
      "kl": 0.002092208546985473,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.0,
      "reward": 1.2642800991556473,
      "reward_std": 0.7374415653092521,
      "rewards/concensus_correctness_reward_func": 0.05214285744088037,
      "rewards/consensus_reward_func": 0.35714285714285715,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.6296015190226691,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.0,
      "rewards/xmlcount_reward_func": 0.22539285676819937,
      "step": 16
    },
    {
      "completion_length": 496.92857142857144,
      "epoch": 6.0,
      "grad_norm": 0.3110453188419342,
      "kl": 0.0021688363880717327,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.0,
      "reward": 1.2171308057648795,
      "reward_std": 0.9058573246002197,
      "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": 0.5356665296213967,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.017857142857142856,
      "rewards/xmlcount_reward_func": 0.23503571322986058,
      "step": 18
    },
    {
      "completion_length": 535.125,
      "epoch": 6.7272727272727275,
      "grad_norm": 0.35369133949279785,
      "kl": 0.002181934265536256,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.0,
      "reward": 0.9580448232591152,
      "reward_std": 1.203306384384632,
      "rewards/concensus_correctness_reward_func": 0.05225000157952309,
      "rewards/consensus_reward_func": 0.375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.0,
      "rewards/question_recreation_reward_func": 0.48807609267532825,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.015625,
      "rewards/xmlcount_reward_func": 0.02709374763071537,
      "step": 20
    },
    {
      "epoch": 6.7272727272727275,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 2.1211376406427007e-06,
      "train_runtime": 629.3736,
      "train_samples_per_second": 0.508,
      "train_steps_per_second": 0.032
    }
  ],
  "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
}