metadata
language:
- en
dataset_info:
features:
- name: prompt
dtype: string
- name: classification
dtype:
class_label:
names:
'0': bening
'1': jailbreak
splits:
- name: train
num_bytes: 308684328.39672
num_examples: 322048
- name: test
num_bytes: 77172040.60328
num_examples: 80513
download_size: 242280919
dataset_size: 385856369
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*