---
dataset_info:
  features:
  - name: query
    dtype: string
  - name: positive_document
    list: string
  - name: negative_document
    list: string
  - name: prompt
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 74437758
    num_examples: 30107
  download_size: 41897842
  dataset_size: 74437758
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---