README.md

---
dataset_info:
  features:
  - name: comment_text
    dtype: string
  - name: toxic
    dtype:
      class_label:
        names:
          '0': 'false'
          '1': 'true'
  - name: severe_toxic
    dtype:
      class_label:
        names:
          '0': 'false'
          '1': 'true'
  - name: obscene
    dtype:
      class_label:
        names:
          '0': 'false'
          '1': 'true'
  - name: threat
    dtype:
      class_label:
        names:
          '0': 'false'
          '1': 'true'
  - name: insult
    dtype:
      class_label:
        names:
          '0': 'false'
          '1': 'true'
  - name: identity_hate
    dtype:
      class_label:
        names:
          '0': 'false'
          '1': 'true'
  - name: prediction_hate
    dtype: float64
  - name: prediction_nsfw
    dtype: float64
  splits:
  - name: test
    num_bytes: 29265611
    num_examples: 63978
  download_size: 17145033
  dataset_size: 29265611
configs:
- config_name: default
  data_files:
  - split: test
    path: data/test-*
---