{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.26666666666666666,
  "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": 0.7875,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 240.3,
      "completions/mean_length": 483.275,
      "completions/mean_terminated_length": 200.9800018310547,
      "completions/min_length": 428.0,
      "completions/min_terminated_length": 172.0,
      "entropy": 0.44431554432958364,
      "epoch": 0.06666666666666667,
      "frac_reward_zero_std": 0.35,
      "grad_norm": 1.53125,
      "kl": 0.038827025642240186,
      "learning_rate": 9.4375e-06,
      "loss": 0.0020309146493673325,
      "num_tokens": 51530.0,
      "reward": 0.45,
      "reward_std": 0.3909654438495636,
      "rewards/JointRewardFunction/mean": 0.45,
      "rewards/JointRewardFunction/std": 0.3909654557704926,
      "step": 10,
      "step_time": 21.882423204801306
    },
    {
      "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": 0.8,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 386.5,
      "completions/mean_length": 502.575,
      "completions/mean_terminated_length": 372.61000061035156,
      "completions/min_length": 457.8,
      "completions/min_terminated_length": 355.4,
      "entropy": 0.259922884311527,
      "epoch": 0.13333333333333333,
      "frac_reward_zero_std": 0.35,
      "grad_norm": 2.890625,
      "kl": 0.0678297683596611,
      "learning_rate": 8.8125e-06,
      "loss": 0.0040422692894935604,
      "num_tokens": 105496.0,
      "reward": 0.575,
      "reward_std": 0.4570506900548935,
      "rewards/JointRewardFunction/mean": 0.575,
      "rewards/JointRewardFunction/std": 0.45705069601535797,
      "step": 20,
      "step_time": 22.13749420610038
    },
    {
      "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": 0.5875,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 422.9,
      "completions/mean_length": 473.825,
      "completions/mean_terminated_length": 379.08262634277344,
      "completions/min_length": 386.8,
      "completions/min_terminated_length": 335.6,
      "entropy": 0.25187050821259616,
      "epoch": 0.2,
      "frac_reward_zero_std": 0.35,
      "grad_norm": 0.09423828125,
      "kl": 0.11119869479443878,
      "learning_rate": 8.1875e-06,
      "loss": 0.0015535765327513219,
      "num_tokens": 154818.0,
      "reward": 0.7875,
      "reward_std": 0.3947398692369461,
      "rewards/JointRewardFunction/mean": 0.7875,
      "rewards/JointRewardFunction/std": 0.39473988115787506,
      "step": 30,
      "step_time": 21.91056177379942
    },
    {
      "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": 0.825,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 429.5,
      "completions/mean_length": 496.7,
      "completions/mean_terminated_length": 396.95,
      "completions/min_length": 420.2,
      "completions/min_terminated_length": 369.0,
      "entropy": 0.24834579518064856,
      "epoch": 0.26666666666666666,
      "frac_reward_zero_std": 0.45,
      "grad_norm": 2.140625,
      "kl": 0.08241275593172759,
      "learning_rate": 7.5625e-06,
      "loss": 0.007686867564916611,
      "num_tokens": 207866.0,
      "reward": 0.6625,
      "reward_std": 0.41830068826675415,
      "rewards/JointRewardFunction/mean": 0.6625,
      "rewards/JointRewardFunction/std": 0.41830070614814757,
      "step": 40,
      "step_time": 21.8902738727018
    }
  ],
  "logging_steps": 10,
  "max_steps": 160,
  "num_input_tokens_seen": 207866,
  "num_train_epochs": 2,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}