---
dataset_info:
  features:
  - name: original
    dtype: string
  - name: corrupted
    dtype: string
  splits:
  - name: train
    num_bytes: 29467122
    num_examples: 56711
  download_size: 14065928
  dataset_size: 29467122
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---