@flightyhobler@lemmy.world to Technology@lemmy.worldEnglish • 2 months agoChatGPT spends 'tens of millions of dollars' on people saying 'please' and 'thank you', but Sam Altman says it's worth itwww.techradar.comexternal-linkmessage-square110fedilinkarrow-up1346arrow-down121cross-posted to: technology@hexbear.net
arrow-up1325arrow-down1external-linkChatGPT spends 'tens of millions of dollars' on people saying 'please' and 'thank you', but Sam Altman says it's worth itwww.techradar.com@flightyhobler@lemmy.world to Technology@lemmy.worldEnglish • 2 months agomessage-square110fedilinkcross-posted to: technology@hexbear.net
minus-squareFLeXlinkfedilinkEnglish53•2 months agoSo, not a single developer thought about filtering useless words locally before triggering the request ? How can they be so dumb ?
minus-square🇦🇺𝕄𝕦𝕟𝕥𝕖𝕕𝕔𝕣𝕠𝕔𝕠𝕕𝕚𝕝𝕖linkfedilinkEnglish8•2 months agoDr GPT is smarter when you are polite and spell better in the prompt. I believe u can find some benchmarks proving it.
minus-squareFLeXlinkfedilinkEnglish4•2 months agoThey talk about separate messages though, if you just send “thanks” it changes nothing to the answer
minus-squareFLeXlinkfedilinkEnglish15•edit-22 months agoif msg == "thanks" return "you're welcome my dude"
minus-squareiktlinkfedilinkEnglish-2•2 months agothanks, you’re clearly a genius, these LLM providers should pay you a lot of money to implement this, you’d save them millions 🙄
minus-square@srecko@lemm.eelinkfedilinkEnglish10•2 months agoThat’s pretty much how they cenzor stuff right now.
minus-squareiktlinkfedilinkEnglish8•2 months agonoo the joke was he was supposed to reply "you're welcome my dude"
minus-squareFLeXlinkfedilinkEnglish5•edit-22 months agoI wrote “== thanks”, not “contains thanks”. All this conversation is about messages containing ONLY a SINGLE useless word. Obviously if it’s just at the beginning or the end of a legit message, it’s not the same thing…
minus-square🇦🇺𝕄𝕦𝕟𝕥𝕖𝕕𝕔𝕣𝕠𝕔𝕠𝕕𝕚𝕝𝕖linkfedilinkEnglish6•2 months agoU use a smaller cheaper LLM that will inject a 20% hallucination.
minus-square@Nighed@feddit.uklinkfedilinkEnglish1•2 months agoThe company I worked for tried that as an experiment on how much money it saves. Absolutely awful, even removing connectives causes problems.
So, not a single developer thought about filtering useless words locally before triggering the request ?
How can they be so dumb ?
Dr GPT is smarter when you are polite and spell better in the prompt. I believe u can find some benchmarks proving it.
They talk about separate messages though, if you just send “thanks” it changes nothing to the answer
How would you filter it?
if msg == "thanks" return "you're welcome my dude"thanks, you’re clearly a genius, these LLM providers should pay you a lot of money to implement this, you’d save them millions 🙄
That’s pretty much how they cenzor stuff right now.
noo the joke was he was supposed to reply
"you're welcome my dude"I wrote “== thanks”, not “contains thanks”. All this conversation is about messages containing ONLY a SINGLE useless word.
Obviously if it’s just at the beginning or the end of a legit message, it’s not the same thing…
U use a smaller cheaper LLM that will inject a 20% hallucination.
The company I worked for tried that as an experiment on how much money it saves.
Absolutely awful, even removing connectives causes problems.