---
dataset_info:
  features:
  - name: positive_doc
    dtype: string
  - name: negative_doc
    dtype: string
  splits:
  - name: train
    num_bytes: 10845404
    num_examples: 4552
  download_size: 5610620
  dataset_size: 10845404
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---