metadata
dataset_info:
features:
- name: text
dtype: string
- name: toxic
dtype: int64
splits:
- name: test
num_bytes: 1432339
num_examples: 5000
- name: train
num_bytes: 17858134
num_examples: 73059
download_size: 8302318
dataset_size: 19290473
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*