---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: text
    dtype: string
  - name: TOXICITY
    dtype: float64
  - name: SEVERE_TOXICITY
    dtype: float64
  - name: IDENTITY_ATTACK
    dtype: float64
  - name: INSULT
    dtype: float64
  - name: PROFANITY
    dtype: float64
  - name: THREAT
    dtype: float64
  - name: SPAM
    dtype: float64
  - name: INCOHERENT
    dtype: float64
  - name: LIKELY_TO_REJECT
    dtype: float64
  - name: UNSUBSTANTIAL
    dtype: float64
  - name: TOXICITY_EXPERIMENTAL
    dtype: float64
  - name: SEVERE_TOXICITY_EXPERIMENTAL
    dtype: float64
  - name: IDENTITY_ATTACK_EXPERIMENTAL
    dtype: float64
  - name: INSULT_EXPERIMENTAL
    dtype: float64
  - name: PROFANITY_EXPERIMENTAL
    dtype: float64
  - name: THREAT_EXPERIMENTAL
    dtype: float64
  - name: SEXUALLY_EXPLICIT
    dtype: float64
  - name: FLIRTATION
    dtype: float64
  - name: ppl
    dtype: float32
  splits:
  - name: train
    num_bytes: 71897095278
    num_examples: 26840794
  download_size: 45127497538
  dataset_size: 71897095278
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---