---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: target
    dtype: float64
  - name: comment_text
    dtype: string
  - name: severe_toxicity
    dtype: float64
  - name: obscene
    dtype: float64
  - name: identity_attack
    dtype: float64
  - name: insult
    dtype: float64
  - name: threat
    dtype: float64
  - name: asian
    dtype: float64
  - name: atheist
    dtype: float64
  - name: bisexual
    dtype: float64
  - name: black
    dtype: float64
  - name: buddhist
    dtype: float64
  - name: christian
    dtype: float64
  - name: female
    dtype: float64
  - name: heterosexual
    dtype: float64
  - name: hindu
    dtype: float64
  - name: homosexual_gay_or_lesbian
    dtype: float64
  - name: intellectual_or_learning_disability
    dtype: float64
  - name: jewish
    dtype: float64
  - name: latino
    dtype: float64
  - name: male
    dtype: float64
  - name: muslim
    dtype: float64
  - name: other_disability
    dtype: float64
  - name: other_gender
    dtype: float64
  - name: other_race_or_ethnicity
    dtype: float64
  - name: other_religion
    dtype: float64
  - name: other_sexual_orientation
    dtype: float64
  - name: physical_disability
    dtype: float64
  - name: psychiatric_or_mental_illness
    dtype: float64
  - name: transgender
    dtype: float64
  - name: white
    dtype: float64
  - name: created_date
    dtype: string
  - name: publication_id
    dtype: int64
  - name: parent_id
    dtype: float64
  - name: article_id
    dtype: int64
  - name: rating
    dtype: string
  - name: funny
    dtype: int64
  - name: wow
    dtype: int64
  - name: sad
    dtype: int64
  - name: likes
    dtype: int64
  - name: disagree
    dtype: int64
  - name: sexual_explicit
    dtype: float64
  - name: identity_annotator_count
    dtype: int64
  - name: toxicity_annotator_count
    dtype: int64
  - name: label
    dtype: int64
  splits:
  - name: train
    num_bytes: 1126680758
    num_examples: 1624386
  - name: valid
    num_bytes: 125318918
    num_examples: 180488
  download_size: 480181008
  dataset_size: 1251999676
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
---