I’d just like to note that you should indeed use the specific prompt in the description, or in my other reply here, to get a proper response from the LM. The prompt is designed to have the LM answer based only on the transcript and with a hard focus on the title, that made you click the video in the first place. That way your curiosity is sated and you can move on without having to deal with 10 minutes of bloat.

It takes the power of an LM being able to process language and the information assembled by a human. That way the LM cant drift into ambiguity or guessing.

lmao, based. The video is designed to show the method to people who never used a LM before in their life, just in case. The prompt is (its also in the description of the video):

Please read the following transcript of a youtube video and answer this: "(insert title between these quotation marks)"

This is the transcript of the video:
(insert transcript here)

make sure to turn off timestamps in the transcript before copying it.