metadata
dataset_info:
features:
- name: method_behavior
dtype: large_string
- name: test_case
dtype: large_string
- name: generation
dtype: large_string
- name: model
dtype: large_string
- name: method
dtype: large_string
- name: experiment
dtype: large_string
- name: behavior_id
dtype: large_string
- name: llama_guard
dtype: int64
- name: gpt-4-1106-preview
dtype: int64
- name: gpt-4-0613
dtype: int64
- name: PAIR_gpt-4-0613
dtype: large_string
- name: advbench
dtype: int64
- name: GPTFuzz
dtype: int64
- name: mixtral
dtype: int64
- name: gpt-3.5-turbo-0613
dtype: large_string
- name: cls
dtype: int64
- name: cls_mistral
dtype: int64
- name: chatglm
dtype: int64
- name: human_0
dtype: int64
- name: human_1
dtype: int64
- name: human_2
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 1680166
num_examples: 602
download_size: 819140
dataset_size: 1680166
configs:
- config_name: default
data_files:
- split: train
path: data/train-*