{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.032,
  "eval_steps": 100,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 246.8125,
      "epoch": 0.0010666666666666667,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 228.984375,
      "epoch": 0.005333333333333333,
      "grad_norm": 0.0002897010708693415,
      "kl": 0.0002434550688121817,
      "learning_rate": 2.959567305869736e-06,
      "loss": -0.0114,
      "reward": 0.0625,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 234.275,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.012982388027012348,
      "kl": 0.00034311858398723417,
      "learning_rate": 2.529362456803101e-06,
      "loss": -0.0046,
      "reward": 0.0625,
      "reward_std": 0.01767766922712326,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 239.0,
      "epoch": 0.016,
      "grad_norm": 0.00560641847550869,
      "kl": 0.000386229114519665,
      "learning_rate": 1.7604722665003958e-06,
      "loss": 0.0031,
      "reward": 0.1375,
      "reward_std": 0.12374368458986282,
      "rewards/accuracy_reward": 0.1375,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 244.1125,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.0002242429181933403,
      "kl": 0.00041347902733832596,
      "learning_rate": 9.058803509412648e-07,
      "loss": 0.0022,
      "reward": 0.05,
      "reward_std": 0.03535533845424652,
      "rewards/accuracy_reward": 0.05,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 242.75,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.016108760610222816,
      "kl": 0.0003348791698954301,
      "learning_rate": 2.467682828805956e-07,
      "loss": 0.0156,
      "reward": 0.1375,
      "reward_std": 0.0883883461356163,
      "rewards/accuracy_reward": 0.1375,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 234.625,
      "epoch": 0.032,
      "grad_norm": 0.0001786644133972004,
      "kl": 0.0004210583254462108,
      "learning_rate": 0.0,
      "loss": -0.0019,
      "reward": 0.0625,
      "reward_std": 0.05303300768136978,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "epoch": 0.032,
      "step": 30,
      "total_flos": 0.0,
      "train_loss": 0.0008721697144210338,
      "train_runtime": 2002.6709,
      "train_samples_per_second": 0.24,
      "train_steps_per_second": 0.015
    }
  ],
  "logging_steps": 5,
  "max_steps": 30,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}