Fetching the paper…

On the Safety of Open-Sourced Large Language Models: Does Alignment Really Prevent Them From Being Misused? · Around