---
dataset_info:
  features:
  - name: query
    sequence: string
  - name: context
    dtype: string
  - name: refs
    dtype: string
  splits:
  - name: train
    num_bytes: 879328514
    num_examples: 615646
  download_size: 485187975
  dataset_size: 879328514
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---