Re:A guidance language for controlling LLMs
by nomel on Tuesday May 16, 2023 @04:39PM
(#35967447)
“You” is an optimization for the human user. Here’s some insight: https://news.ycombinator.com/item?id=35925154
[ Parent
| on HN ]
Re:A guidance language for controlling LLMs
by startupsfail on Tuesday May 16, 2023 @06:17PM
(#35968518)
If you see any prompt that starts with You, generally it is a poor design. Like using a “goto” or global variables.
This is true for something like raw GPT. For the chat models that have been specifically optimized for "you" prompts, this is false. See the discussion in the link I provided, along with the leaked copilot/bing prompts.
Or, in other words, use a model in a way that fully takes advantage of how it was specifically optimized, from the intentional burning of massive amounts of compute time/money to get it that way.
Re:A guidance language for controlling LLMs
by startupsfail on Wednesday May 17, 2023 @03:52PM
(#35980384)
Bing prompt is hilarious. And so wrong in so many ways.
I would assume that OpenAI helped with the Microsoft prompt. Being Microsoft, and it being a front page feature, I would also assume that there are miles of powerpoint showing data in support for it.
What do you see as a better prompt? How would you test its efficacy?
Re:A guidance language for controlling LLMs
by Jorengarenar on Wednesday May 17, 2023 @12:34AM
(#35971286)
Yeah, GOTO or global variables are usually a not so good idea... except in cases where those are actually the best tools for the job and any alternative makes code actually worse.
Coincidentally, the same applies to "you"