{
  "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": 277.15625,
      "epoch": 0.5714285714285714,
      "grad_norm": 5.865261554718018,
      "kl": 11.93383551388979,
      "learning_rate": 4.965903258506806e-07,
      "loss": 0.0119,
      "reward": 8.516574621200562,
      "reward_std": 5.593167632818222,
      "rewards/concensus_correctness_reward_func": 5.221750069409609,
      "rewards/consensus_reward_func": 0.9375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 0.9375,
      "rewards/question_recreation_reward_func": 0.5981057807803154,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.203125,
      "rewards/xmlcount_reward_func": 0.6185937505215406,
      "step": 2
    },
    {
      "completion_length": 249.0,
      "epoch": 1.0,
      "grad_norm": 1.4555927515029907,
      "kl": 2.5505317747592926,
      "learning_rate": 4.698684378016222e-07,
      "loss": 0.0019,
      "reward": 20.225919882456463,
      "reward_std": 9.441229522228241,
      "rewards/concensus_correctness_reward_func": 15.145333290100098,
      "rewards/consensus_reward_func": 1.5,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.8333333333333333,
      "rewards/question_recreation_reward_func": 0.7291280776262283,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.22916666666666666,
      "rewards/xmlcount_reward_func": 0.7889583458503088,
      "step": 4
    },
    {
      "completion_length": 266.03125,
      "epoch": 1.5714285714285714,
      "grad_norm": 2.5842504501342773,
      "kl": 1.9852429926395416,
      "learning_rate": 4.193203929064353e-07,
      "loss": 0.002,
      "reward": 11.658353358507156,
      "reward_std": 4.756189459934831,
      "rewards/concensus_correctness_reward_func": 7.802937511354685,
      "rewards/consensus_reward_func": 1.3125,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.125,
      "rewards/question_recreation_reward_func": 0.6328532956540585,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.171875,
      "rewards/xmlcount_reward_func": 0.613187488168478,
      "step": 6
    },
    {
      "completion_length": 253.29166666666666,
      "epoch": 2.0,
      "grad_norm": 2.71340274810791,
      "kl": 3.834065690636635,
      "learning_rate": 3.5042385616324236e-07,
      "loss": 0.0029,
      "reward": 16.266943613688152,
      "reward_std": 9.872911716500917,
      "rewards/concensus_correctness_reward_func": 12.076083501180014,
      "rewards/consensus_reward_func": 1.1666666666666667,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.5833333333333333,
      "rewards/question_recreation_reward_func": 0.6288601706425349,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.20833333333333334,
      "rewards/xmlcount_reward_func": 0.603666678071022,
      "step": 8
    },
    {
      "completion_length": 253.84375,
      "epoch": 2.571428571428571,
      "grad_norm": 2.902672052383423,
      "kl": 14.077829856425524,
      "learning_rate": 2.706448363680831e-07,
      "loss": 0.0141,
      "reward": 11.773723363876343,
      "reward_std": 6.151409147307277,
      "rewards/concensus_correctness_reward_func": 7.921500042080879,
      "rewards/consensus_reward_func": 0.9375,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.25,
      "rewards/question_recreation_reward_func": 0.6172233410179615,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.203125,
      "rewards/xmlcount_reward_func": 0.8443750068545341,
      "step": 10
    },
    {
      "completion_length": 239.33333333333334,
      "epoch": 3.0,
      "grad_norm": 1.6412651538848877,
      "kl": 11.135655641555786,
      "learning_rate": 1.886286282148002e-07,
      "loss": 0.0084,
      "reward": 10.80202571551005,
      "reward_std": 7.497294177611669,
      "rewards/concensus_correctness_reward_func": 6.923333485921224,
      "rewards/consensus_reward_func": 0.9166666666666666,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.4166666666666667,
      "rewards/question_recreation_reward_func": 0.5331921527783076,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.1875,
      "rewards/xmlcount_reward_func": 0.8246666689713796,
      "step": 12
    },
    {
      "completion_length": 221.9375,
      "epoch": 3.571428571428571,
      "grad_norm": 2.7689168453216553,
      "kl": 7.665247939527035,
      "learning_rate": 1.1326296046939333e-07,
      "loss": 0.0077,
      "reward": 14.8109350502491,
      "reward_std": 9.853634387254715,
      "rewards/concensus_correctness_reward_func": 10.454687546938658,
      "rewards/consensus_reward_func": 1.1875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.3125,
      "rewards/question_recreation_reward_func": 0.6400289982557297,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.265625,
      "rewards/xmlcount_reward_func": 0.9505937620997429,
      "step": 14
    },
    {
      "completion_length": 262.8333333333333,
      "epoch": 4.0,
      "grad_norm": 10.040779113769531,
      "kl": 8.907942593097687,
      "learning_rate": 5.271487265090163e-08,
      "loss": 0.0067,
      "reward": 11.699872573216757,
      "reward_std": 7.398251553376515,
      "rewards/concensus_correctness_reward_func": 7.810000052054723,
      "rewards/consensus_reward_func": 1.3333333333333333,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.25,
      "rewards/question_recreation_reward_func": 0.5266643911600113,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.14583333333333334,
      "rewards/xmlcount_reward_func": 0.6340416620175043,
      "step": 16
    },
    {
      "completion_length": 261.75,
      "epoch": 4.571428571428571,
      "grad_norm": 3.726018190383911,
      "kl": 38.97571674361825,
      "learning_rate": 1.3545689574841341e-08,
      "loss": 0.039,
      "reward": 14.649300217628479,
      "reward_std": 10.099224492907524,
      "rewards/concensus_correctness_reward_func": 10.412562608718872,
      "rewards/consensus_reward_func": 1.1875,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.4375,
      "rewards/question_recreation_reward_func": 0.7462379038333893,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.203125,
      "rewards/xmlcount_reward_func": 0.6623749986756593,
      "step": 18
    },
    {
      "completion_length": 264.2916666666667,
      "epoch": 5.0,
      "grad_norm": 1.6413745880126953,
      "kl": 19.3692533771197,
      "learning_rate": 0.0,
      "loss": 0.0145,
      "reward": 12.260410388310751,
      "reward_std": 7.397667676210403,
      "rewards/concensus_correctness_reward_func": 7.908166567484538,
      "rewards/consensus_reward_func": 1.3333333333333333,
      "rewards/cumulative_reward_2": 0.0,
      "rewards/final_correctness_reward_func": 1.3333333333333333,
      "rewards/question_recreation_reward_func": 0.593618909517924,
      "rewards/soft_format_reward_func": 0.0,
      "rewards/strict_format_reward_func": 0.2916666666666667,
      "rewards/xmlcount_reward_func": 0.8002916822830836,
      "step": 20
    },
    {
      "epoch": 5.0,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 0.0108985282946378,
      "train_runtime": 247.4117,
      "train_samples_per_second": 1.293,
      "train_steps_per_second": 0.081
    }
  ],
  "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": 4,
  "trial_name": null,
  "trial_params": null
}