---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: prompt
    dtype: string
  - name: answers
    struct:
    - name: text
      sequence: string
    - name: answer_start
      sequence: int64
  splits:
  - name: train
    num_bytes: 55324690
    num_examples: 11873
  - name: validation
    num_bytes: 55324690
    num_examples: 11873
  download_size: 12946390
  dataset_size: 110649380
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---