metadata
dataset_info:
features:
- name: text
dtype: string
- name: reason
dtype: string
- name: class
dtype: string
- name: toxic_level
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 130517
num_examples: 1049
download_size: 54543
dataset_size: 130517
configs:
- config_name: default
data_files:
- split: train
path: data/train-*