←back to thread

Uncensor any LLM with abliteration

(huggingface.co)
586 points mizzao | 1 comments | 13 Jun 24 03:42 UTC | HN request time: 0.335s | source
Show context
vasco ◴[13 Jun 24 06:49 UTC] No.40666684[source]▶
> "As an AI assistant, I cannot help you." While this safety feature is crucial for preventing misuse,

What is the safety added by this? What is unsafe about a computer giving you answers?

replies(11): >>40666709 #>>40666828 #>>40666835 #>>40666890 #>>40666984 #>>40666992 #>>40667025 #>>40667243 #>>40667633 #>>40669842 #>>40670809 #
1. FeepingCreature ◴[13 Jun 24 07:08 UTC] No.40666835[source]▶
People keep claiming they can publish weights and also prevent misuse, such as spam and, a bit later on, stuff like helping people build bombs.

This is of course impossible, but that makes certain companies' approaches unviable, so they keep claiming it anyways.