---
dataset_info:
  features:
  - name: original_text
    dtype: string
  - name: detoxified_text
    dtype: string
  - name: context
    dtype: string
  - name: source
    dtype: string
  - name: filename
    dtype: string
  splits:
  - name: train
    num_bytes: 16755914
    num_examples: 36800
  download_size: 5625151
  dataset_size: 16755914
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---