---
dataset_info:
  features:
  - name: corrupted
    dtype: string
  - name: correct
    dtype: string
  splits:
  - name: train
    num_bytes: 76914
    num_examples: 1376
  download_size: 32048
  dataset_size: 76914
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---