---
dataset_info:
  features:
  - name: category
    dtype: string
  - name: source
    dtype: string
  - name: text
    dtype: string
  - name: corrupted_text
    dtype: string
  - name: score
    dtype: float64
  - name: tags
    sequence: int64
  splits:
  - name: train
    num_bytes: 65201521
    num_examples: 79832
  - name: test
    num_bytes: 8149957
    num_examples: 10000
  download_size: 92193012
  dataset_size: 73351478
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---