{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.992,
  "eval_steps": 500,
  "global_step": 31,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 777.4896087646484,
      "epoch": 0.032,
      "grad_norm": 0.5903529116709957,
      "kl": 0.0,
      "learning_rate": 7.5e-07,
      "loss": 0.0,
      "reward": 0.5625000223517418,
      "reward_std": 0.07392781227827072,
      "rewards/accuracy_reward": 0.5625000223517418,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "completion_length": 777.8633041381836,
      "epoch": 0.16,
      "grad_norm": 0.380767083039185,
      "kl": 0.0002472996711730957,
      "learning_rate": 2.989857536612915e-06,
      "loss": 0.0,
      "reward": 0.5247395876795053,
      "reward_std": 0.03600824438035488,
      "rewards/accuracy_reward": 0.5247395876795053,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "completion_length": 808.4958587646485,
      "epoch": 0.32,
      "grad_norm": 0.17029855265794955,
      "kl": 0.0011264801025390625,
      "learning_rate": 2.649066664678467e-06,
      "loss": 0.0,
      "reward": 0.48437500894069674,
      "reward_std": 0.04581574220210314,
      "rewards/accuracy_reward": 0.48437500894069674,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "completion_length": 786.0593933105469,
      "epoch": 0.48,
      "grad_norm": 3.440906400450222,
      "kl": 0.018999862670898437,
      "learning_rate": 1.9302048490666355e-06,
      "loss": 0.0008,
      "reward": 0.47604167386889457,
      "reward_std": 0.039101397059857844,
      "rewards/accuracy_reward": 0.47604167386889457,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "completion_length": 777.956265258789,
      "epoch": 0.64,
      "grad_norm": 0.3120699062992745,
      "kl": 0.006772041320800781,
      "learning_rate": 1.069795150933365e-06,
      "loss": 0.0003,
      "reward": 0.47083334177732467,
      "reward_std": 0.029002375528216363,
      "rewards/accuracy_reward": 0.47083334177732467,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 810.5562713623046,
      "epoch": 0.8,
      "grad_norm": 0.2740538627520351,
      "kl": 0.003619384765625,
      "learning_rate": 3.5093333532153313e-07,
      "loss": 0.0001,
      "reward": 0.49375000819563863,
      "reward_std": 0.03593750298023224,
      "rewards/accuracy_reward": 0.49375000819563863,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "completion_length": 798.4125213623047,
      "epoch": 0.96,
      "grad_norm": 9.502030258134457,
      "kl": 0.008373069763183593,
      "learning_rate": 1.0142463387085465e-08,
      "loss": 0.0003,
      "reward": 0.5312500089406967,
      "reward_std": 0.05033150520175696,
      "rewards/accuracy_reward": 0.5312500089406967,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "completion_length": 795.6406402587891,
      "epoch": 0.992,
      "kl": 0.006107330322265625,
      "reward": 0.5885416716337204,
      "reward_std": 0.012757758609950542,
      "rewards/accuracy_reward": 0.5885416716337204,
      "rewards/format_reward": 0.0,
      "step": 31,
      "total_flos": 0.0,
      "train_loss": 0.00025426253455163744,
      "train_runtime": 3666.8179,
      "train_samples_per_second": 0.273,
      "train_steps_per_second": 0.008
    }
  ],
  "logging_steps": 5,
  "max_steps": 31,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "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": 16,
  "trial_name": null,
  "trial_params": null
}