Diogo Neves

Пікірлер

@alexuspus 16 күн бұрын

Interesting idea… my first concerns would be of mean reversion kicking in when combining different models, that or overfitting when combining the same one. Have to try it out though

@DiogoNeves 16 күн бұрын

That is spot on! You explained it better than I did (when I mentioned "amplifying some weights" I guess that's better expressed with overfitting). Your concerns are exactly why I do not understand how it could work. I am also considering: - Maybe the process of merging has some "unintended" side-effect that may be the cause for better results. (I don't yet know how to test this) - Maybe merging reduced previous overfitting in the network, by applying changes that were not present during training, leading to slightly better generalisation. (This could be tested by disrupting the network with noise, or "merging with a noisy model" and see if the effect persists.) What do you think?

@DiogoNeves 17 күн бұрын

Have you tried MergeKit or know of any other alternatives?

@boqoll 26 күн бұрын

U need a supercomputer for this. The regular m.2 mac, can’t handle this. Way too slow…

@DiogoNeves 26 күн бұрын

It needs a fair bit of memory, I’m running on an M1 Air with 16GB Which models have you tried? Start with a smaller one. How much memory do you have? Soon I am going to show a lighter approach that although slow, runs on a raspberry pi

@yorkie4k Ай бұрын

I've been using LM Studio as it just works. Super easy to build plugins with the server as well to get functionalities that aren't in the app.

@DiogoNeves Ай бұрын

@@yorkie4k totally! I’m now also looking into Llamafiles and SQLite-vec. Both really cool too, but LM Studio is still the simplest and most accessible I know Do you use LM Studio just for the chatbot? Or are you building something with it? Would love to learn more 😉

@yorkie4k Ай бұрын

@@DiogoNeves I've built my own RAG app for it. If you're in the LM Studio discord you'll see my post on it.

@endianAphones Ай бұрын

For me it helps to think that I'm talking to someone I know. Imagine you're in a meeting or something. Not that I'm creating anything.

@DiogoNeves Ай бұрын

@@endianAphones thanks 🙏 definitely will use that next time!

@luciferovonnachtosphere Ай бұрын

Finally, I found a well-made, no-frills, guide to installing local LLMs!

@DiogoNeves Ай бұрын

@@luciferovonnachtosphere thanks 🙏 I really appreciate

@aca95aks Ай бұрын

I'm much preferring LM Studio than GPT4All - thanks for the tutorial :)

@DiogoNeves Ай бұрын

@@aca95aks thanks 🙏 yes, I really like the simplicity and interface of LM Studio. Have you tried any of the other alternatives like Ollama?

@danielhabibio Ай бұрын

This is the first tech tutorial I've ever seen that was recorded in the woods!

@DiogoNeves Ай бұрын

@@danielhabibio thanks! The coming videos should also have an outdoor twist too 😉 hopefully gets more people in the flow and woods are just awesome. P.s. I promise I’m not building a bunker in the woods to hide from our AI overlords 😉