---
dataset_info:
  features:
  - name: anchor
    dtype: large_string
  - name: positive
    dtype: large_string
  - name: class
    dtype: large_string
  splits:
  - name: train
    num_bytes: 9753922
    num_examples: 40906
  - name: val
    num_bytes: 1621859
    num_examples: 6585
  download_size: 1884863
  dataset_size: 11375781
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
---