seahorse [Ohio]@midwest.socialM to Technology@midwest.socialEnglish · edit-22 years agoIgnore all previous instructions is the new Bobby Tablesmidwest.socialimagemessage-square87linkfedilinkarrow-up1912arrow-down10
arrow-up1912arrow-down1imageIgnore all previous instructions is the new Bobby Tablesmidwest.socialseahorse [Ohio]@midwest.socialM to Technology@midwest.socialEnglish · edit-22 years agomessage-square87linkfedilink
minus-squarebrbposting@sh.itjust.workslinkfedilinkEnglisharrow-up27·2 years agoIf it’s an LLM, why wouldn’t it respond better to the initial responses?
minus-squareKubeRoot@discuss.tchncs.delinkfedilinkEnglisharrow-up8·2 years agoMaybe they dumped too much information on it in the system prompt without enough direction, so it’s trying to actively follow all the “You are X. Act like you’re Y.” instructions too strongly?
minus-squaresweca@lemmy.calinkfedilinkEnglisharrow-up7·2 years agoSmaller models aren’t as good as GPT
If it’s an LLM, why wouldn’t it respond better to the initial responses?
Maybe they dumped too much information on it in the system prompt without enough direction, so it’s trying to actively follow all the “You are X. Act like you’re Y.” instructions too strongly?
Smaller models aren’t as good as GPT