---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: metadata
    struct:
    - name: difficulty
      struct:
      - name: num_characteristics
        dtype: int64
      - name: num_people
        dtype: int64
    - name: source_dataset
      dtype: string
    - name: source_index
      dtype: int64
  splits:
  - name: train
    num_bytes: 34429202
    num_examples: 20000
  download_size: 7317983
  dataset_size: 34429202
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---