---
dataset_info:
  features:
  - name: corrupted_text
    dtype: string
  - name: corrected_text
    dtype: string
  splits:
  - name: train
    num_bytes: 94628
    num_examples: 1039
  download_size: 49480
  dataset_size: 94628
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---