5 ms·
He's talking about llama 2 superhots, and mistral derivatives that can be uncensored
by senseiV 3y ago
He's talking about llama 2 superhots, and mistral derivatives that can be uncensored
- mvdtnz 3y agoI literally have no idea what you just said.
- pests 3y agoLlama2 and minstal are models Derivatives are dervied from those base models Unsensored being obvious I hope Superhots are a derived model (for llama only? Not sure) using a certain technique for fine tuning and context length. A Google will reveal the paper and research.
- mvdtnz 3y agoOk this doesn't answer the question. What are these terrible things people are doing with these models?
- Pigalowda 3y agoEdgelording
- TheCleric 3y agoI prefer the term “pizza cutter”: all edge and no point.
- xbmcuser 3y agoThe op never said anything about someone doing something terrible just that looking at these models and what is possible large corps or government controlling what llm models will do is a pipe dream so author of the original article does not need to worry about it.
- mvdtnz 3y agoSorry, not "terrible", but "shocking". Presumably not shockingly good and wholesome.
- hackerlight 3y agoNot LLMs but I've seen multiple cases of neo-Nazi propaganda made with image generators and shared on social media. The concern with LLM misuse is probably similar to this. LLMs could also be used by adversaries as part of their campaign to sow civic discord using social media. What they could do is use the LLM to create thousands of extremist fake personas that amplify extremist ideas on both sides to try to disrupt cohesion and unity, and win the strategic rivalry without having to fire a gun. One angle people aren't considering is that that restraining LLMs is our first alignment project. Putting aside whether we should try to restrain LLMs, it's an interesting question in itself (regarding safety) if we technically struggle to do that.