David Gerard@awful.systemscakeM to TechTakes@awful.systemsEnglish · 1 年前a collection of thousands of prompts attempting to control the idiot genie of ChatGPTgithub.comexternal-linkmessage-square7linkfedilinkarrow-up146arrow-down10
arrow-up146arrow-down1external-linka collection of thousands of prompts attempting to control the idiot genie of ChatGPTgithub.comDavid Gerard@awful.systemscakeM to TechTakes@awful.systemsEnglish · 1 年前message-square7linkfedilink
minus-squarebitfucker@programming.devlinkfedilinkEnglisharrow-up6·edit-21 年前Nothing? I’m pointing out another way to “attack” some LLM?
minus-squareShakingMyHead@awful.systemslinkfedilinkEnglisharrow-up9·1 年前Probably would have been easier when the context window wasn’t 128k. Though what the point would be should someone actually achieve that eludes me a bit.
minus-squarebitfucker@programming.devlinkfedilinkEnglisharrow-up10·1 年前I don’t know man. Maybe a chat bot for help center that forgot they shouldn’t give refund or non existent promotion?
minus-squarenfultz@awful.systemslinkfedilinkEnglisharrow-up1·1 年前I have LLMs translate novels, I can tell when they run over the window because they forget characters’ gender half way through, things like that. Lots of legal documents are even longer, maybe.
and your point is?
Nothing? I’m pointing out another way to “attack” some LLM?
Probably would have been easier when the context window wasn’t 128k.
Though what the point would be should someone actually achieve that eludes me a bit.
I don’t know man. Maybe a chat bot for help center that forgot they shouldn’t give refund or non existent promotion?
deleted by creator
I have LLMs translate novels, I can tell when they run over the window because they forget characters’ gender half way through, things like that. Lots of legal documents are even longer, maybe.