I bet this works because of the way insta and fb work.

How about we just put them to bed once in a while?

Please elaborate on this one

I think they mean that the model should have sleep period where they update themselves with what they learnt that day.

Why not learn how to catch lobsters?

Is anyone excited to do ablative testing on it?

With such a high throughput because of sparsity, I'm particulary interested in distilling it into other architectures. I'd like to try a recurrent transformer when I have the time

Don't vampires sleep in the day? I read the tech manual, no personal experience, ahem.

Oh I wonder how dating works.

... normally? they don't have the same "30% of adults will never marry because of arbitrary bullshit" that modern/western countries have.

Is that using the WHIP output or something else?

Yep WHIP output!

* PR - https://github.com/obsproject/obs-studio/pull/7926

* RFC - https://datatracker.ietf.org/doc/rfc9725/

He wasn't as productive as he could have been. This seems like Chuck Norris and Jeff Dean territory after all.

In small companies there is one lane.

Which lane was I in?

Back to the drawing board. What about a proximity sensor?

I think what I want to do, is have a dodgy local LLM that picks up the context that the user is speaking to the LLM, and then enables it for 20 minutes or so.

But even thats a bit of a wild tradeoff.