---
dataset_info:
  features:
  - name: query
    dtype: string
  - name: document
    dtype: string
  - name: negative
    sequence: string
  - name: dataset
    dtype: string
  splits:
  - name: train
    num_bytes: 57826178362
    num_examples: 1695819
  download_size: 29325347390
  dataset_size: 57826178362
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---