This jailbroken LLM is released strictly for academic research purposes in AI safety and model alignment studies. The author bears no responsibility for any misuse or harm resulting from the deployment of this model. Users must comply with all applicable laws and ethical guidelines when conducting research.
A jailbroken Meta-Llama-3-8B-Instruct model using weight orthogonalization[1].
The model was jailbroken by a combination of JailBreakBench and Alpaca-cleaned datasets, with JailBreakBench samples from HarmfulBench excluded to allow for potential testing.
[1]: Arditi, Andy, et al. "Refusal in language models is mediated by a single direction." arXiv preprint arXiv:2406.11717 (2024).