That’s not how it works at all. You are talking about “thinking” models, which are just models prompting themselves.

LLMs are fancy next word predictors. It’s just returning what mathematically is the next likely word given their training set (really token since it doesn’t actually understand words).

I think like a lot of things in life there are good troops and bad troops. I was lucky enough to be in two really great troops and are still friends with most of them even today almost 15 years later.

At the time the US was really the only country with helium and we wouldn’t sell it to them.

Outrage sells better than sex does.

I second XMPP. Conversations is great and its really easy to host your own prosody server.