{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6666666666666666,
  "eval_steps": 500,
  "global_step": 100,
  "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": 0.7375,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 284.9,
      "completions/mean_length": 491.525,
      "completions/mean_terminated_length": 261.77666931152345,
      "completions/min_length": 438.2,
      "completions/min_terminated_length": 233.4,
      "entropy": 0.23946007741615177,
      "epoch": 0.06666666666666667,
      "frac_reward_zero_std": 0.45,
      "grad_norm": 1.453125,
      "kl": 0.05540826761134667,
      "learning_rate": 9.100000000000001e-06,
      "loss": 0.0011166661977767944,
      "num_tokens": 52190.0,
      "reward": 0.7,
      "reward_std": 0.35391277372837066,
      "rewards/JointRewardFunction/mean": 0.7,
      "rewards/JointRewardFunction/std": 0.3539127916097641,
      "step": 10,
      "step_time": 22.393758795900045
    },
    {
      "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": 0.7625,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 380.1,
      "completions/mean_length": 495.3625,
      "completions/mean_terminated_length": 357.5166717529297,
      "completions/min_length": 439.9,
      "completions/min_terminated_length": 337.5,
      "entropy": 0.15637785932049156,
      "epoch": 0.13333333333333333,
      "frac_reward_zero_std": 0.5,
      "grad_norm": 0.024169921875,
      "kl": 0.07800487959757448,
      "learning_rate": 8.1e-06,
      "loss": -0.0069251880049705505,
      "num_tokens": 105579.0,
      "reward": 0.775,
      "reward_std": 0.3446744382381439,
      "rewards/JointRewardFunction/mean": 0.775,
      "rewards/JointRewardFunction/std": 0.3446744501590729,
      "step": 20,
      "step_time": 22.42510473759985
    },
    {
      "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": 0.8125,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 382.0,
      "completions/mean_length": 500.8125,
      "completions/mean_terminated_length": 368.27000122070314,
      "completions/min_length": 451.0,
      "completions/min_terminated_length": 348.6,
      "entropy": 0.16634994922205806,
      "epoch": 0.2,
      "frac_reward_zero_std": 0.65,
      "grad_norm": 1.5390625,
      "kl": 0.10815342329442501,
      "learning_rate": 7.100000000000001e-06,
      "loss": 0.005232430994510651,
      "num_tokens": 157060.0,
      "reward": 0.8875,
      "reward_std": 0.2394672751426697,
      "rewards/JointRewardFunction/mean": 0.8875,
      "rewards/JointRewardFunction/std": 0.23946728110313414,
      "step": 30,
      "step_time": 22.14527411999982
    },
    {
      "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": 0.8875,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 262.4,
      "completions/mean_length": 502.0125,
      "completions/mean_terminated_length": 256.325,
      "completions/min_length": 451.3,
      "completions/min_terminated_length": 246.5,
      "entropy": 0.15448929518461227,
      "epoch": 0.26666666666666666,
      "frac_reward_zero_std": 0.45,
      "grad_norm": 0.02001953125,
      "kl": 0.08631132342852652,
      "learning_rate": 6.1e-06,
      "loss": 0.0021702755242586137,
      "num_tokens": 210533.0,
      "reward": 0.7625,
      "reward_std": 0.3475900113582611,
      "rewards/JointRewardFunction/mean": 0.7625,
      "rewards/JointRewardFunction/std": 0.34759002923965454,
      "step": 40,
      "step_time": 22.401593531500158
    },
    {
      "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": 0.925,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 235.5,
      "completions/mean_length": 509.125,
      "completions/mean_terminated_length": 234.8,
      "completions/min_length": 490.1,
      "completions/min_terminated_length": 234.1,
      "entropy": 0.15993599854409696,
      "epoch": 0.3333333333333333,
      "frac_reward_zero_std": 0.6,
      "grad_norm": 1.3125,
      "kl": 0.10001510812435299,
      "learning_rate": 5.1e-06,
      "loss": -0.001868080161511898,
      "num_tokens": 263979.0,
      "reward": 0.8375,
      "reward_std": 0.29291952252388,
      "rewards/JointRewardFunction/mean": 0.8375,
      "rewards/JointRewardFunction/std": 0.2929195284843445,
      "step": 50,
      "step_time": 22.454355105500007
    },
    {
      "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": 0.9125,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 222.0,
      "completions/mean_length": 505.65,
      "completions/mean_terminated_length": 219.93333435058594,
      "completions/min_length": 474.0,
      "completions/min_terminated_length": 218.0,
      "entropy": 0.15155332698486745,
      "epoch": 0.4,
      "frac_reward_zero_std": 0.65,
      "grad_norm": 1.265625,
      "kl": 0.0995997670572251,
      "learning_rate": 4.1e-06,
      "loss": 0.0030021272599697114,
      "num_tokens": 317831.0,
      "reward": 0.8125,
      "reward_std": 0.32402919232845306,
      "rewards/JointRewardFunction/mean": 0.8125,
      "rewards/JointRewardFunction/std": 0.32402920424938203,
      "step": 60,
      "step_time": 22.315176106599846
    },
    {
      "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": 0.9125,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 182.4,
      "completions/mean_length": 505.3875,
      "completions/mean_terminated_length": 174.23333435058595,
      "completions/min_length": 475.2,
      "completions/min_terminated_length": 168.0,
      "entropy": 0.13972499519586562,
      "epoch": 0.4666666666666667,
      "frac_reward_zero_std": 0.65,
      "grad_norm": 0.0235595703125,
      "kl": 0.11641144650056959,
      "learning_rate": 3.1000000000000004e-06,
      "loss": 0.002840310521423817,
      "num_tokens": 372058.0,
      "reward": 0.8,
      "reward_std": 0.3038551896810532,
      "rewards/JointRewardFunction/mean": 0.8,
      "rewards/JointRewardFunction/std": 0.30385519564151764,
      "step": 70,
      "step_time": 22.520917878499766
    },
    {
      "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": 0.8875,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 238.4,
      "completions/mean_length": 504.55,
      "completions/mean_terminated_length": 225.6,
      "completions/min_length": 466.8,
      "completions/min_terminated_length": 210.8,
      "entropy": 0.14867580346763135,
      "epoch": 0.5333333333333333,
      "frac_reward_zero_std": 0.7,
      "grad_norm": 1.109375,
      "kl": 0.09885827025864273,
      "learning_rate": 2.1000000000000002e-06,
      "loss": 9.710401645861566e-05,
      "num_tokens": 426094.0,
      "reward": 0.8625,
      "reward_std": 0.2558668524026871,
      "rewards/JointRewardFunction/mean": 0.8625,
      "rewards/JointRewardFunction/std": 0.25586686432361605,
      "step": 80,
      "step_time": 22.384258029100057
    },
    {
      "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": 0.9375,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 144.1,
      "completions/mean_length": 508.4875,
      "completions/mean_terminated_length": 139.3,
      "completions/min_length": 492.9,
      "completions/min_terminated_length": 134.5,
      "entropy": 0.1391168820671737,
      "epoch": 0.6,
      "frac_reward_zero_std": 0.65,
      "grad_norm": 1.1953125,
      "kl": 0.11537287631072105,
      "learning_rate": 1.1e-06,
      "loss": -0.004437017068266868,
      "num_tokens": 480921.0,
      "reward": 0.7875,
      "reward_std": 0.29291952252388,
      "rewards/JointRewardFunction/mean": 0.7875,
      "rewards/JointRewardFunction/std": 0.2929195284843445,
      "step": 90,
      "step_time": 22.70512801439986
    },
    {
      "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": 0.9125,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 230.5,
      "completions/mean_length": 504.3625,
      "completions/mean_terminated_length": 220.3,
      "completions/min_length": 466.1,
      "completions/min_terminated_length": 210.1,
      "entropy": 0.1445184069685638,
      "epoch": 0.6666666666666666,
      "frac_reward_zero_std": 0.75,
      "grad_norm": 0.0107421875,
      "kl": 0.09200373792555183,
      "learning_rate": 1.0000000000000001e-07,
      "loss": 0.002389365620911121,
      "num_tokens": 536154.0,
      "reward": 0.8625,
      "reward_std": 0.22220884561538695,
      "rewards/JointRewardFunction/mean": 0.8625,
      "rewards/JointRewardFunction/std": 0.22220885157585143,
      "step": 100,
      "step_time": 22.48313269170003
    }
  ],
  "logging_steps": 10,
  "max_steps": 100,
  "num_input_tokens_seen": 536154,
  "num_train_epochs": 1,
  "save_steps": 20,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}