{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8275862068965516,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 750.0,
      "completions/max_terminated_length": 750.0,
      "completions/mean_length": 208.6875,
      "completions/mean_terminated_length": 208.6875,
      "completions/min_length": 19.5,
      "completions/min_terminated_length": 19.5,
      "epoch": 0.27586206896551724,
      "grad_norm": 18.433101654052734,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": -0.1616,
      "num_tokens": 7435.0,
      "reward": 0.05778068071231246,
      "reward_std": 0.013260316103696823,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "rewards/concensus_correctness_reward_func/std": 0.0,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.05778068397194147,
      "rewards/question_recreation_reward_func/std": 0.03348450642079115,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.0,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 2
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 415.5,
      "completions/max_terminated_length": 415.5,
      "completions/mean_length": 130.5625,
      "completions/mean_terminated_length": 130.5625,
      "completions/min_length": 13.5,
      "completions/min_terminated_length": 13.5,
      "epoch": 0.5517241379310345,
      "grad_norm": 8.727158546447754,
      "kl": 0.0008754075024626218,
      "learning_rate": 4.864543104251586e-07,
      "loss": 0.0809,
      "num_tokens": 13620.0,
      "reward": 0.10532870702445507,
      "reward_std": 0.024424075847491622,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "rewards/concensus_correctness_reward_func/std": 0.0,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.10532871540635824,
      "rewards/question_recreation_reward_func/std": 0.03869475517421961,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.0,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 4
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0625,
      "completions/max_length": 669.5,
      "completions/max_terminated_length": 507.5,
      "completions/mean_length": 312.6875,
      "completions/mean_terminated_length": 273.9107208251953,
      "completions/min_length": 44.0,
      "completions/min_terminated_length": 44.0,
      "epoch": 0.8275862068965517,
      "grad_norm": 10.41586971282959,
      "kl": 0.0009672067608335055,
      "learning_rate": 4.472851273490984e-07,
      "loss": 0.2478,
      "num_tokens": 22719.0,
      "reward": 0.2098427265882492,
      "reward_std": 0.2130617438815534,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "rewards/concensus_correctness_reward_func/std": 0.0,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.125,
      "rewards/final_correctness_reward_func/std": 0.3535533845424652,
      "rewards/question_recreation_reward_func/mean": 0.07696773111820221,
      "rewards/question_recreation_reward_func/std": 0.04010230861604214,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.00787500012665987,
      "rewards/xmlcount_reward_func/std": 0.022273864597082138,
      "step": 6
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.125,
      "completions/max_length": 873.5,
      "completions/max_terminated_length": 632.0,
      "completions/mean_length": 329.5,
      "completions/mean_terminated_length": 224.16666412353516,
      "completions/min_length": 30.5,
      "completions/min_terminated_length": 30.5,
      "epoch": 1.1379310344827587,
      "grad_norm": 4.472732067108154,
      "kl": 0.0007184029036579886,
      "learning_rate": 3.867370395306068e-07,
      "loss": 0.0869,
      "num_tokens": 32190.0,
      "reward": 0.19183985888957977,
      "reward_std": 0.2473740577697754,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "rewards/concensus_correctness_reward_func/std": 0.0,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.125,
      "rewards/final_correctness_reward_func/std": 0.3535533845424652,
      "rewards/question_recreation_reward_func/mean": 0.019964855164289474,
      "rewards/question_recreation_reward_func/std": 0.012113618198782206,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.046875,
      "rewards/xmlcount_reward_func/std": 0.13258251547813416,
      "step": 8
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 718.5,
      "completions/max_terminated_length": 718.5,
      "completions/mean_length": 255.0,
      "completions/mean_terminated_length": 255.0,
      "completions/min_length": 6.0,
      "completions/min_terminated_length": 6.0,
      "epoch": 1.4137931034482758,
      "grad_norm": 14.328417778015137,
      "kl": 0.001265099534066394,
      "learning_rate": 3.1137137178519977e-07,
      "loss": 0.182,
      "num_tokens": 40366.0,
      "reward": 1.4011108474805951,
      "reward_std": 1.9615169735625386,
      "rewards/concensus_correctness_reward_func/mean": 1.25,
      "rewards/concensus_correctness_reward_func/std": 3.535534143447876,
      "rewards/consensus_reward_func/mean": 0.125,
      "rewards/consensus_reward_func/std": 0.3535533845424652,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.0182358892634511,
      "rewards/question_recreation_reward_func/std": 0.014001987874507904,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.00787500012665987,
      "rewards/xmlcount_reward_func/std": 0.022273864597082138,
      "step": 10
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 522.5,
      "completions/max_terminated_length": 522.5,
      "completions/mean_length": 172.0,
      "completions/mean_terminated_length": 172.0,
      "completions/min_length": 7.5,
      "completions/min_terminated_length": 7.5,
      "epoch": 1.6896551724137931,
      "grad_norm": 8.893263816833496,
      "kl": 0.0011968070393777452,
      "learning_rate": 2.2935516363191693e-07,
      "loss": 0.3234,
      "num_tokens": 47214.0,
      "reward": 0.023529402911663055,
      "reward_std": 0.014957319013774395,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "rewards/concensus_correctness_reward_func/std": 0.0,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.02352940198034048,
      "rewards/question_recreation_reward_func/std": 0.020326449535787106,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.0,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 12
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 614.0,
      "completions/max_terminated_length": 614.0,
      "completions/mean_length": 354.4000015258789,
      "completions/mean_terminated_length": 354.4000015258789,
      "completions/min_length": 96.5,
      "completions/min_terminated_length": 96.5,
      "epoch": 1.9655172413793105,
      "grad_norm": 7.406893730163574,
      "kl": 0.0008268929668702185,
      "learning_rate": 1.4957614383675767e-07,
      "loss": 0.1587,
      "num_tokens": 57685.0,
      "reward": 0.01602007821202278,
      "reward_std": 0.007509211776778102,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "rewards/concensus_correctness_reward_func/std": 0.0,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.016020076349377632,
      "rewards/question_recreation_reward_func/std": 0.009688148740679026,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.0,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 14
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0625,
      "completions/max_length": 659.5,
      "completions/max_terminated_length": 355.0,
      "completions/mean_length": 197.125,
      "completions/mean_terminated_length": 141.28571319580078,
      "completions/min_length": 7.5,
      "completions/min_terminated_length": 7.5,
      "epoch": 2.2758620689655173,
      "grad_norm": 9.953104972839355,
      "kl": 0.0009395094821229577,
      "learning_rate": 8.067960709356478e-08,
      "loss": 0.1676,
      "num_tokens": 64935.0,
      "reward": 0.2408544309437275,
      "reward_std": 0.3167252861894667,
      "rewards/concensus_correctness_reward_func/mean": 0.21975000202655792,
      "rewards/concensus_correctness_reward_func/std": 0.6215468645095825,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.02110444661229849,
      "rewards/question_recreation_reward_func/std": 0.01698118494823575,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.0,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 16
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0625,
      "completions/max_length": 743.5,
      "completions/max_terminated_length": 467.0,
      "completions/mean_length": 253.125,
      "completions/mean_terminated_length": 204.83928680419922,
      "completions/min_length": 15.0,
      "completions/min_terminated_length": 15.0,
      "epoch": 2.5517241379310347,
      "grad_norm": 7.2750020027160645,
      "kl": 0.0010990903319907375,
      "learning_rate": 3.013156219837776e-08,
      "loss": 0.369,
      "num_tokens": 73081.0,
      "reward": 0.02212073840200901,
      "reward_std": 0.015308489557355642,
      "rewards/concensus_correctness_reward_func/mean": 0.0,
      "rewards/concensus_correctness_reward_func/std": 0.0,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.022120739333331585,
      "rewards/question_recreation_reward_func/std": 0.0167312603443861,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.0,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 18
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 358.0,
      "completions/max_terminated_length": 358.0,
      "completions/mean_length": 135.75,
      "completions/mean_terminated_length": 135.75,
      "completions/min_length": 13.5,
      "completions/min_terminated_length": 13.5,
      "epoch": 2.8275862068965516,
      "grad_norm": 15.697056770324707,
      "kl": 0.001014976151054725,
      "learning_rate": 3.4096741493194193e-09,
      "loss": 0.2045,
      "num_tokens": 79349.0,
      "reward": 0.24373424425721169,
      "reward_std": 0.3253984763287008,
      "rewards/concensus_correctness_reward_func/mean": 0.22374999523162842,
      "rewards/concensus_correctness_reward_func/std": 0.6328606009483337,
      "rewards/consensus_reward_func/mean": 0.0,
      "rewards/consensus_reward_func/std": 0.0,
      "rewards/cumulative_reward_2/mean": 0.0,
      "rewards/cumulative_reward_2/std": 0.0,
      "rewards/final_correctness_reward_func/mean": 0.0,
      "rewards/final_correctness_reward_func/std": 0.0,
      "rewards/question_recreation_reward_func/mean": 0.019984250888228416,
      "rewards/question_recreation_reward_func/std": 0.009346078149974346,
      "rewards/soft_format_reward_func/mean": 0.0,
      "rewards/soft_format_reward_func/std": 0.0,
      "rewards/strict_format_reward_func/mean": 0.0,
      "rewards/strict_format_reward_func/std": 0.0,
      "rewards/xmlcount_reward_func/mean": 0.0,
      "rewards/xmlcount_reward_func/std": 0.0,
      "step": 20
    },
    {
      "epoch": 2.8275862068965516,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 0.16590909510850907,
      "train_runtime": 553528.733,
      "train_samples_per_second": 0.0,
      "train_steps_per_second": 0.0
    }
  ],
  "logging_steps": 2,
  "max_steps": 20,
  "num_input_tokens_seen": 79349,
  "num_train_epochs": 3,
  "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
}