ekZepp@lemmy.world to Technology@lemmy.mlEnglish · 7 months agoStudy Finds That 52 Percent of ChatGPT Answers to Programming Questions Are Wrongfuturism.comexternal-linkmessage-square131fedilinkarrow-up1631arrow-down118
arrow-up1613arrow-down1external-linkStudy Finds That 52 Percent of ChatGPT Answers to Programming Questions Are Wrongfuturism.comekZepp@lemmy.world to Technology@lemmy.mlEnglish · 7 months agomessage-square131fedilink
minus-squareWalnutLum@lemmy.mllinkfedilinkarrow-up7·7 months agoThis is because all LLMs function primarily based on the token context you feed it. The best way to use any LLM is to completely fill up it’s history with relevant context, then ask your question.
minus-squareBoozilla@lemmy.worldlinkfedilinkEnglisharrow-up4·7 months agoI worked on a creative writing thing with it and the more I added, the better its responses. And 4 is a noticeable improvement over 3.5.
This is because all LLMs function primarily based on the token context you feed it.
The best way to use any LLM is to completely fill up it’s history with relevant context, then ask your question.
I worked on a creative writing thing with it and the more I added, the better its responses. And 4 is a noticeable improvement over 3.5.