---
dataset_info:
  features:
  - name: anchor
    dtype: string
  - name: pos_neg
    dtype: string
  - name: label
    dtype: int64
  splits:
  - name: train
    num_bytes: 674717980
    num_examples: 1095846
  download_size: 160768607
  dataset_size: 674717980
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---