---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: ground_truth
    dtype: string
  - name: initial_responses
    sequence: string
  - name: initial_trimmed_token_lists
    sequence:
      sequence: int64
  splits:
  - name: train
    num_bytes: 95254767
    num_examples: 198
  download_size: 14724746
  dataset_size: 95254767
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---