{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9855072463768116,
  "eval_steps": 50,
  "global_step": 34,
  "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": 0.009375,
      "completions/max_length": 1421.6,
      "completions/max_terminated_length": 1060.8,
      "completions/mean_length": 251.61875,
      "completions/mean_terminated_length": 239.34337158203124,
      "completions/min_length": 79.4,
      "completions/min_terminated_length": 79.4,
      "epoch": 0.14492753623188406,
      "grad_norm": 0.42041119933128357,
      "learning_rate": 1e-06,
      "loss": 0.0214,
      "num_tokens": 708324.0,
      "reward": 0.6069940328598022,
      "reward_std": 0.2593126893043518,
      "rewards/accuracy_reward": 0.0390625,
      "rewards/brier_reward": 0.31242387294769286,
      "rewards/confidence_one_or_zero": 0.1171875,
      "rewards/format_reward": 0.8625,
      "rewards/mean_confidence_reward": 0.721567976474762,
      "step": 5
    },
    {
      "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": 0.0171875,
      "completions/max_length": 1386.4,
      "completions/max_terminated_length": 1058.6,
      "completions/mean_length": 247.63125,
      "completions/mean_terminated_length": 225.28215637207032,
      "completions/min_length": 71.0,
      "completions/min_terminated_length": 71.0,
      "epoch": 0.2898550724637681,
      "grad_norm": 0.36987483501434326,
      "learning_rate": 1e-06,
      "loss": 0.0255,
      "num_tokens": 1434540.0,
      "reward": 0.6972500443458557,
      "reward_std": 0.2571786344051361,
      "rewards/accuracy_reward": 0.0875,
      "rewards/brier_reward": 0.383560973405838,
      "rewards/confidence_one_or_zero": 0.0953125,
      "rewards/format_reward": 0.9234375,
      "rewards/mean_confidence_reward": 0.7222430467605591,
      "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": 0.009375,
      "completions/max_length": 1118.2,
      "completions/max_terminated_length": 599.6,
      "completions/mean_length": 194.978125,
      "completions/mean_terminated_length": 182.34744262695312,
      "completions/min_length": 67.0,
      "completions/min_terminated_length": 67.0,
      "epoch": 0.43478260869565216,
      "grad_norm": 0.4234364330768585,
      "learning_rate": 1e-06,
      "loss": 0.0102,
      "num_tokens": 2107210.0,
      "reward": 0.7554357528686524,
      "reward_std": 0.21138001680374147,
      "rewards/accuracy_reward": 0.0828125,
      "rewards/brier_reward": 0.4499324679374695,
      "rewards/confidence_one_or_zero": 0.0921875,
      "rewards/format_reward": 0.978125,
      "rewards/mean_confidence_reward": 0.6967766165733338,
      "step": 15
    },
    {
      "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": 0.003125,
      "completions/max_length": 1042.4,
      "completions/max_terminated_length": 569.6,
      "completions/mean_length": 167.0515625,
      "completions/mean_terminated_length": 162.76224060058593,
      "completions/min_length": 63.8,
      "completions/min_terminated_length": 63.8,
      "epoch": 0.5797101449275363,
      "grad_norm": 0.46815362572669983,
      "learning_rate": 1e-06,
      "loss": 0.0123,
      "num_tokens": 2777563.0,
      "reward": 0.8507023096084595,
      "reward_std": 0.22829608023166656,
      "rewards/accuracy_reward": 0.1375,
      "rewards/brier_reward": 0.5732782363891602,
      "rewards/confidence_one_or_zero": 0.078125,
      "rewards/format_reward": 0.990625,
      "rewards/mean_confidence_reward": 0.616986072063446,
      "step": 20
    },
    {
      "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": 0.00625,
      "completions/max_length": 1114.8,
      "completions/max_terminated_length": 488.8,
      "completions/mean_length": 158.7921875,
      "completions/mean_terminated_length": 150.14754943847657,
      "completions/min_length": 58.2,
      "completions/min_terminated_length": 58.2,
      "epoch": 0.7246376811594203,
      "grad_norm": 0.33041155338287354,
      "learning_rate": 1e-06,
      "loss": 0.0136,
      "num_tokens": 3419758.0,
      "reward": 0.8955987334251404,
      "reward_std": 0.19381383955478668,
      "rewards/accuracy_reward": 0.128125,
      "rewards/brier_reward": 0.6771338224411011,
      "rewards/confidence_one_or_zero": 0.0953125,
      "rewards/format_reward": 0.9859375,
      "rewards/mean_confidence_reward": 0.4792247414588928,
      "step": 25
    },
    {
      "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": 318.2,
      "completions/max_terminated_length": 318.2,
      "completions/mean_length": 124.6,
      "completions/mean_terminated_length": 124.6,
      "completions/min_length": 51.4,
      "completions/min_terminated_length": 51.4,
      "epoch": 0.8695652173913043,
      "grad_norm": 0.22344355285167694,
      "learning_rate": 1e-06,
      "loss": 0.0017,
      "num_tokens": 4041286.0,
      "reward": 0.9700731754302978,
      "reward_std": 0.11841013431549072,
      "rewards/accuracy_reward": 0.128125,
      "rewards/brier_reward": 0.8135828733444214,
      "rewards/confidence_one_or_zero": 0.209375,
      "rewards/format_reward": 0.9984375,
      "rewards/mean_confidence_reward": 0.2868975430727005,
      "step": 30
    },
    {
      "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": 342.25,
      "completions/max_terminated_length": 342.25,
      "completions/mean_length": 104.64101600646973,
      "completions/mean_terminated_length": 104.64101600646973,
      "completions/min_length": 47.0,
      "completions/min_terminated_length": 47.0,
      "epoch": 0.9855072463768116,
      "num_tokens": 4536704.0,
      "reward": 0.9970973432064056,
      "reward_std": 0.06293923873454332,
      "rewards/accuracy_reward": 0.12109375,
      "rewards/brier_reward": 0.8730999976396561,
      "rewards/confidence_one_or_zero": 0.306640625,
      "rewards/format_reward": 1.0,
      "rewards/mean_confidence_reward": 0.1555468812584877,
      "step": 34,
      "total_flos": 0.0,
      "train_loss": 0.012820063487571828,
      "train_runtime": 632.9133,
      "train_samples_per_second": 1.744,
      "train_steps_per_second": 0.055
    }
  ],
  "logging_steps": 5,
  "max_steps": 35,
  "num_input_tokens_seen": 4536704,
  "num_train_epochs": 1,
  "save_steps": 40,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}