---
dataset_info:
  features:
  - name: query
    dtype: string
  - name: positive
    sequence: string
  - name: negative
    sequence: string
  - name: cluster
    dtype: int64
  splits:
  - name: train
    num_bytes: 40055781
    num_examples: 18906
  download_size: 832801
  dataset_size: 40055781
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---