{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.761904761904762,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 999.6105346679688,
      "epoch": 0.38095238095238093,
      "grad_norm": 154.4364013671875,
      "kl": 0.0,
      "learning_rate": 3e-06,
      "loss": -0.019,
      "reward": 0.7500000298023224,
      "reward_std": 0.3680399917066097,
      "rewards/accuracy_reward": 0.031250001629814506,
      "rewards/format_reward": 0.7187500298023224,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1006.7606468200684,
      "epoch": 2.380952380952381,
      "grad_norm": 8.609855651855469,
      "kl": 1.3177490234375,
      "learning_rate": 1.7604722665003958e-06,
      "loss": 0.0393,
      "reward": 0.7695312835276127,
      "reward_std": 0.3738137981854379,
      "rewards/accuracy_reward": 0.04101562706637196,
      "rewards/format_reward": 0.7285156585276127,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1006.158299255371,
      "epoch": 4.761904761904762,
      "grad_norm": 0.9559445381164551,
      "kl": 2.4314453125,
      "learning_rate": 0.0,
      "loss": 0.0869,
      "reward": 0.7937500372529029,
      "reward_std": 0.36111804023385047,
      "rewards/accuracy_reward": 0.03906250209547579,
      "rewards/format_reward": 0.7546875327825546,
      "step": 10
    },
    {
      "epoch": 4.761904761904762,
      "step": 10,
      "total_flos": 0.0,
      "train_loss": 0.057270023971796036,
      "train_runtime": 3066.7214,
      "train_samples_per_second": 0.548,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 5,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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
}