I haven’t tried neural-chat, but the combined model seems to be better (anecdotally) than OH2.5/Mistral at following instructions, reasoning, some of the overall quirks with llama.cpp seem to be ironed out with it too.

My critical thinking professor told an anecdote:

“This is a pen; how do we know? Because it does everything that a pen does. And that’s good enough for now, but when the pen gets up and starts talking, it’s time to reevaluate whether it’s still a pen.”

I think that applies here too. We are the sum of our experiences and education is a part of that experience, eventually we have enough knowledge in a particular subject and a concept for gaining new knowledge. So, to a certain extent it never happens, we should always be asking questions.

Didn’t they murder a journalist too?

I call it: stupid and proud

I definitely don’t agree with their opinions, and I think it would be unconstitutional as hell to implement the measures that they’re advocating for, but it should be noted that if they are successful, we’ll see such a spectacular torrenting and dark web LLM scene. I don’t think there’s any stopping this. They can try and they will lose.

You’re all beta testers for Rocky Linux.