metadata
dataset_info:
features:
- name: text
dtype: string
- name: label
dtype:
class_label:
names:
'0': not-toxic
'1': toxic
splits:
- name: train
num_bytes: 1709552
num_examples: 16800
download_size: 1137292
dataset_size: 1709552
configs:
- config_name: default
data_files:
- split: train
path: data/train-*