{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.975609756097561,
  "eval_steps": 500,
  "global_step": 40,
  "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": 0.9958333333333333,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 88.4,
      "completions/mean_length": 255.9635437011719,
      "completions/mean_terminated_length": 88.4,
      "completions/min_length": 242.0,
      "completions/min_terminated_length": 88.4,
      "epoch": 0.24390243902439024,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.3228670060634613,
      "kl": 0.0005886403494514525,
      "learning_rate": 1.3802524773477098e-06,
      "loss": 0.0001,
      "num_tokens": 540050.0,
      "reward": 0.33145859837532043,
      "reward_std": 0.026524317637085914,
      "rewards/wrapper/mean": 0.165729296207428,
      "rewards/wrapper/std": 0.019212284684181215,
      "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": 0.9958333333333333,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 84.6,
      "completions/mean_length": 255.95364685058593,
      "completions/mean_terminated_length": 84.6,
      "completions/min_length": 238.2,
      "completions/min_terminated_length": 84.6,
      "epoch": 0.4878048780487805,
      "frac_reward_zero_std": 0.0010416666977107526,
      "grad_norm": 0.2965202331542969,
      "kl": 0.0011322355625452475,
      "learning_rate": 3.105568074032347e-06,
      "loss": 0.0001,
      "num_tokens": 1079737.0,
      "reward": 0.3324328601360321,
      "reward_std": 0.028421235084533692,
      "rewards/wrapper/mean": 0.16621642410755158,
      "rewards/wrapper/std": 0.019593635573983192,
      "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": 1.0,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 256.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 256.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.7317073170731707,
      "frac_reward_zero_std": 0.0031250000931322573,
      "grad_norm": 0.284412145614624,
      "kl": 0.0018224636791273952,
      "learning_rate": 4.830883670716985e-06,
      "loss": 0.0001,
      "num_tokens": 1618911.0,
      "reward": 0.3351516366004944,
      "reward_std": 0.027410299330949784,
      "rewards/wrapper/mean": 0.1675758183002472,
      "rewards/wrapper/std": 0.019275682047009468,
      "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": 0.9979166666666668,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 40.6,
      "completions/mean_length": 255.97239685058594,
      "completions/mean_terminated_length": 40.6,
      "completions/min_length": 245.4,
      "completions/min_terminated_length": 40.6,
      "epoch": 0.975609756097561,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.3010239601135254,
      "kl": 0.003081864328123629,
      "learning_rate": 6.556199267401621e-06,
      "loss": 0.0002,
      "num_tokens": 2158306.0,
      "reward": 0.3369597256183624,
      "reward_std": 0.029058777168393134,
      "rewards/wrapper/mean": 0.16847985982894897,
      "rewards/wrapper/std": 0.01961274929344654,
      "step": 20
    },
    {
      "epoch": 0.975609756097561,
      "eval_clip_ratio/high_max": 0.0,
      "eval_clip_ratio/high_mean": 0.0,
      "eval_clip_ratio/low_mean": 0.0,
      "eval_clip_ratio/low_min": 0.0,
      "eval_clip_ratio/region_mean": 0.0,
      "eval_completions/clipped_ratio": 1.0,
      "eval_completions/max_length": 256.0,
      "eval_completions/max_terminated_length": 0.0,
      "eval_completions/mean_length": 256.0,
      "eval_completions/mean_terminated_length": 0.0,
      "eval_completions/min_length": 256.0,
      "eval_completions/min_terminated_length": 0.0,
      "eval_frac_reward_zero_std": 0.0,
      "eval_kl": 0.004834831831976772,
      "eval_loss": 0.00019452218839433044,
      "eval_num_tokens": 2158306.0,
      "eval_reward": 0.3390903663635254,
      "eval_reward_std": 0.025529487021267415,
      "eval_rewards/wrapper/mean": 0.1695451831817627,
      "eval_rewards/wrapper/std": 0.017574026770889758,
      "eval_runtime": 76.535,
      "eval_samples_per_second": 2.613,
      "eval_steps_per_second": 0.17,
      "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": 1.0,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 256.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 256.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 1.2439024390243902,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.3230460584163666,
      "kl": 0.007101990515366197,
      "learning_rate": 8.281514864086258e-06,
      "loss": 0.0003,
      "num_tokens": 2698886.0,
      "reward": 0.34189478754997255,
      "reward_std": 0.02718120105564594,
      "rewards/wrapper/mean": 0.1709473878145218,
      "rewards/wrapper/std": 0.019510553032159806,
      "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": 1.0,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 256.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 256.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 1.4878048780487805,
      "frac_reward_zero_std": 0.002083333395421505,
      "grad_norm": 0.29992207884788513,
      "kl": 0.015912340488284826,
      "learning_rate": 1.0006830460770896e-05,
      "loss": 0.0006,
      "num_tokens": 3238296.0,
      "reward": 0.34575957655906675,
      "reward_std": 0.028136394917964935,
      "rewards/wrapper/mean": 0.17287977933883666,
      "rewards/wrapper/std": 0.019339724257588387,
      "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": 1.0,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 256.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 256.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 1.7317073170731707,
      "frac_reward_zero_std": 0.0010416666977107526,
      "grad_norm": 0.2933114469051361,
      "kl": 0.02282416820526123,
      "learning_rate": 1.1732146057455534e-05,
      "loss": 0.0009,
      "num_tokens": 3777860.0,
      "reward": 0.3552273988723755,
      "reward_std": 0.028055840730667116,
      "rewards/wrapper/mean": 0.17761370837688445,
      "rewards/wrapper/std": 0.019311992824077605,
      "step": 35
    },
    {
      "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": 1.0,
      "completions/max_length": 256.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 256.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 256.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 1.975609756097561,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 1.8470395803451538,
      "kl": 0.08914270466193557,
      "learning_rate": 3.884255434752975e-06,
      "loss": 0.0036,
      "num_tokens": 4317276.0,
      "reward": 0.3673787772655487,
      "reward_std": 0.028479888290166854,
      "rewards/wrapper/mean": 0.18368937373161315,
      "rewards/wrapper/std": 0.0193321343511343,
      "step": 40
    },
    {
      "epoch": 1.975609756097561,
      "eval_clip_ratio/high_max": 0.0,
      "eval_clip_ratio/high_mean": 0.0,
      "eval_clip_ratio/low_mean": 0.0,
      "eval_clip_ratio/low_min": 0.0,
      "eval_clip_ratio/region_mean": 0.0,
      "eval_completions/clipped_ratio": 1.0,
      "eval_completions/max_length": 256.0,
      "eval_completions/max_terminated_length": 0.0,
      "eval_completions/mean_length": 256.0,
      "eval_completions/mean_terminated_length": 0.0,
      "eval_completions/min_length": 256.0,
      "eval_completions/min_terminated_length": 0.0,
      "eval_frac_reward_zero_std": 0.0,
      "eval_kl": 0.0393575007468462,
      "eval_loss": 0.0015887623885646462,
      "eval_num_tokens": 4317276.0,
      "eval_reward": 0.3702333927154541,
      "eval_reward_std": 0.026990585755556823,
      "eval_rewards/wrapper/mean": 0.18511669635772704,
      "eval_rewards/wrapper/std": 0.01845288172364235,
      "eval_runtime": 76.3221,
      "eval_samples_per_second": 2.62,
      "eval_steps_per_second": 0.17,
      "step": 40
    }
  ],
  "logging_steps": 5,
  "max_steps": 40,
  "num_input_tokens_seen": 4317276,
  "num_train_epochs": 2,
  "save_steps": 500,
  "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": 24,
  "trial_name": null,
  "trial_params": null
}