---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: target
    dtype: string
  splits:
  - name: train
    num_bytes: 11595842
    num_examples: 20106
  download_size: 5079964
  dataset_size: 11595842
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---