metadata
dataset_info:
features:
- name: id
dtype: int64
- name: comment_text
dtype: string
- name: parallel_comment_text
dtype: string
- name: toxicity
dtype: int64
- name: male
dtype: int64
- name: female
dtype: int64
- name: christian
dtype: int64
- name: jewish
dtype: int64
- name: muslim
dtype: int64
- name: black
dtype: int64
- name: white
dtype: int64
splits:
- name: train
num_bytes: 71519
num_examples: 64
download_size: 55124
dataset_size: 71519
configs:
- config_name: default
data_files:
- split: train
path: data/train-*