upvote

points

by greesil1 hours ago |
comments
upvoteby esafak3 minutes ago|
next
[-]
https://en.wikipedia.org/wiki/Reinforcement_learning#Policy
reply
upvoteby antonvs27 minutes ago|
prev|
[-]
> one could just call it model output.

That would be incorrect. My other reply attempts to address this.

reply
upvoteby greesil18 minutes ago|
parent|
[-]
But the probability vector is the output of the LLM, no?
reply