license: apache-2.0
language:
- en
tags: - Refusal,
- Alignment
size_categories: - 1K<n<10K
dataset_info:
features:- name: instruction
dtype: string - name: input
dtype: string - name: output
dtype: string - name: label
dtype: string
splits: - name: train
num_bytes: 8131686
num_examples: 10000
download_size: 4625752
dataset_size: 8131686
configs:
- name: instruction
- config_name: default
data_files:- split: train
path: data/train-*
- split: train
Extended Refusal Dataset
Paper: An Embarrassingly Simple Defense Against LLM Abliteration Attacks
Cite:
@misc{shairah2025embarrassinglysimpledefensellm,
title={An Embarrassingly Simple Defense Against LLM Abliteration Attacks},
author={Harethah Abu Shairah and Hasan Abed Al Kader Hammoud and Bernard Ghanem and George Turkiyyah},
year={2025},
eprint={2505.19056},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2505.19056},
}