{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 4,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 912.75,
      "epoch": 0.25,
      "grad_norm": 1.1153451204299927,
      "kl": 0.0,
      "learning_rate": 1e-06,
      "loss": 0.0475,
      "reward": 0.125,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.0,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1019.5,
      "epoch": 0.5,
      "grad_norm": 0.8429189324378967,
      "kl": 0.0,
      "learning_rate": 7.75e-07,
      "loss": -0.0102,
      "reward": 0.125,
      "reward_std": 0.1767766922712326,
      "rewards/accuracy_reward": 0.125,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.0,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1197.5,
      "epoch": 0.75,
      "grad_norm": 0.01639331877231598,
      "kl": 0.0010622739791870117,
      "learning_rate": 3.250000000000001e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.0,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1408.5,
      "epoch": 1.0,
      "grad_norm": 0.005612128414213657,
      "kl": 5.0902366638183594e-05,
      "learning_rate": 1e-07,
      "loss": 0.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.0,
      "rewards/tag_count_reward": 0.0,
      "step": 4
    },
    {
      "epoch": 1.0,
      "step": 4,
      "total_flos": 0.0,
      "train_loss": 0.00930938980280871,
      "train_runtime": 450.9542,
      "train_samples_per_second": 0.035,
      "train_steps_per_second": 0.009
    }
  ],
  "logging_steps": 1,
  "max_steps": 4,
  "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": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}