---
dataset_info:
  features:
  - name: query
    dtype: string
  - name: documents
    sequence: string
  - name: scores
    sequence: float64
  splits:
  - name: train
    num_bytes: 5101199648
    num_examples: 109810
  download_size: 2872383380
  dataset_size: 5101199648
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---