---
dataset_info:
  features:
  - name: ref
    dtype: string
  - name: cob
    dtype: string
  splits:
  - name: train
    num_bytes: 11864363
    num_examples: 6618
  download_size: 3511083
  dataset_size: 11864363
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---