---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: comment_text_bpe_tokenize_simple_dup_normalization
    dtype: string
  - name: toxic
    dtype: int64
  - name: severe_toxic
    dtype: int64
  - name: obscene
    dtype: int64
  - name: threat
    dtype: int64
  - name: insult
    dtype: int64
  - name: identity_hate
    dtype: int64
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 60937398
    num_examples: 127656
  - name: validation
    num_bytes: 15066647
    num_examples: 31915
  - name: test
    num_bytes: 29332039
    num_examples: 63978
  download_size: 59838960
  dataset_size: 105336084
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---