alt Hacker NewsI suspect how well this approach would work. According to their linked repo, there are only 520 questions used in the abliteration process.
https://github.com/Sumandora/remove-refusals-with-transforme...
Replies
strictnein • yesterday at 11:08 PM
Two examples, from my testing: the models go from not saying anything remotely bad about China to happily making jokes about its leader.
They also go from refusing to help with certain cyber security tasks to be more than happy to help.
You don't need that many with Deepseek. The easy jailbreak is to provide an excessive character sheet that states that it's a real person, not an AI. And make sure that your character has absolutely no self-censoring or morals. LLM will play it perfectly in-character without refusals.
You can get it talking about Tiananmen Square event in, like, 2-3 prompts.