Nostr Web Client

I am saying nothing about its effects on the economy or how it may affect jobs. Im strictly speaking of the technology itself, and I fail to be impressed with it. Got one on my embassy server and I really dont see all the fuss. As it currently stands, they are truly nothing more than very robust search engines. What they may or may not be in the future is open for speculation, but its still just code and silicon, neither of which have 'intelligence'. They can only do what a human programmed them to do.

mark tyler 2y ago

What model(s) are you using?

I don’t know if it’s important but it seems like it’s important to stress that it’s not a search engine at all. Maybe there are times when they are useful as a search engine, but mostly it’s a thinking engine.

And when i use it as a search engine, I think of it as almost the opposite of robust - I think it behaves more like one person‘s opinion, rather than a summary of what humanity is saying about a topic.

Totally agree that the future is speculation, but also projecting that past trends in capability improvements will continue (especially as it gets exponentially more funding and attention) doesn’t seem too crazy.

One thing search engines can’t do is think with you about something that’s never been asked before. These can.

Reply to this note

Please Login to reply.

Discussion

Joel Walbert 2y ago

Its kinda is a search engine on steroids. It searches data and returns a response. And it can only return results based on data it has access too. Just because the user end gives detailed answers doesnt mean that the backend really does much more than a search engine. Just a much more complex algorithm. As for what model I was trying, not exactly sure. Something with dolphin in the name maybe. Its one that is available with the GPT package that is available for embassy servers

Joel Walbert 2y ago

Ok its called FreeGPT and the specific model is Dolphin-Llama2-7B

mark tyler 2y ago

If that’s the only one that you have messed around with and maybe that could be part of the overpromising vibe you’ve gotten. Llama2-7B is quite a bit less capable than the full llama2 model. And that model is quite a bit less capable than ChatGPT4. Token generation on your embassy is probably a lot slower than what I’m used to using open AI servers.. though self hosted is awesome. I want to get some hardware to do that myself someday.

Joel Walbert 2y ago

I've seen other people's results from the newest before I even touched one myself. Like I said, Im not shitting in it. Only shitting in the blown way out of proportion claims about em. There's nothing magical (for lack of a better term) about em. They do what they do, but there is nothing that can reasonably or rationably be called intelligent about em. When one answers a long sought after question that no human has been able to, then maybe I'll change my tune, but as it currently stands, its just searching data and returning the most likely desired results.

Joel Walbert 2y ago

Oh as for the self hosting, embassy can run on any pc with the proper specs or RasPi so if ya got an old pc or laptop laying around, it may work