Context, then answer... instead of having everything ride on the first character (e.g. we make it pick "Y" or "N" first in response to a yes-or-no question, it usually picks "Y" even if it later talks itself out of it).
Context, then answer... instead of having everything ride on the first character (e.g. we make it pick "Y" or "N" first in response to a yes-or-no question, it usually picks "Y" even if it later talks itself out of it).
7 Comments
BroBot9000@lemmy.world · 11 pts · 248d
Or even better, don’t use the racist pile of linear algebra that regurgitates misinformation and propaganda.
TheLeadenSea@sh.itjust.works · 9 pts · 248d
That's the basis of reasoning models. Make LLMs 'think' through the problem for several hundred tokens before giving a final answer.
saltesc@lemmy.world · 6 pts · 248d
Crank the temperature settings and have it say "Trust me, bro."
Apytele@sh.itjust.works · 2 pts · 247d
Rhynoplaz@lemmy.world · 2 pts · 248d
More wise it would sound.
blimthepixie@lemmy.dbzer0.com · 1 pts · 248d
Remember when satnavs first came out and you could download different voices for them?
klymilark@herbicide.fallcounty.omg.lol · 1 pts · 247d
If only Waze weren't owned by Google... I saw they had that function. Wonder if you can do that with OSMAnd...
DoubleDongle@lemmy.world · 1 pts · 248d
It would definitely be funnier to train it to do that.