User Deleted@lemmy.dbzer0.com to asklemmy@lemmy.mlEnglish · edit-21 year agoDeletedmessage-squaremessage-square178fedilinkarrow-up1100arrow-down13file-text
arrow-up197arrow-down1message-squareDeletedUser Deleted@lemmy.dbzer0.com to asklemmy@lemmy.mlEnglish · edit-21 year agomessage-square178fedilinkfile-text
minus-squarecwagner@discuss.tchncs.delinkfedilinkEnglisharrow-up2·1 year agoGPT 3.5 (what chatGPT was at the beginning) failed at non-trivial math ;) It couldn’t figure out how many characters even were in a word.
minus-squareSirGolan@lemmy.sdf.orglinkfedilinkEnglisharrow-up1·1 year agoYeah. It still definitely does! The interesting thing is that it seems to be very good at estimating and the final answer it gives is usually pretty close to correct in my experience. Of course close doesn’t really count in math problems.
minus-squarecwagner@discuss.tchncs.delinkfedilinkEnglisharrow-up1·1 year agoJust tested it, at least with number of characters in a word and splitting words, GPT4 does it flawlessly.
GPT 3.5 (what chatGPT was at the beginning) failed at non-trivial math ;) It couldn’t figure out how many characters even were in a word.
Yeah. It still definitely does! The interesting thing is that it seems to be very good at estimating and the final answer it gives is usually pretty close to correct in my experience. Of course close doesn’t really count in math problems.
Just tested it, at least with number of characters in a word and splitting words, GPT4 does it flawlessly.