Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Aren't these still essentially completion models under the hood?

If so, my understanding for these preambles is that they need a seed to complete their answer.



But the seed is the user input.


Maybe until the model outputs some affirming preamble, it’s still somewhat probable that it might disagree with the user’s request? So the agreement fluff is kind of like it making the decision to heed the request. Especially if we the consider tokens as the medium by which the model “thinks”. Not to anthropomorphize the damn things too much.

Also I wonder if it could be a side effect of all the supposed alignment efforts that go into training. If you train in a bunch of negative reinforcement samples where the model says something like “sorry I can’t do that” maybe it pushes the model to say things like “sure I’ll do that” in positive cases too?

Disclaimer that I am just yapping




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: