---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: wikipedia
    num_bytes: 79104967
    num_examples: 4000
  - name: train
    num_bytes: 79104967
    num_examples: 4000
  download_size: 73164200
  dataset_size: 158209934
configs:
- config_name: default
  data_files:
  - split: wikipedia
    path: data/wikipedia-*
  - split: train
    path: data/train-*
---