this post was submitted on 07 May 2024
45 points (97.9% liked)

Technology

37800 readers
326 users here now

A nice place to discuss rumors, happenings, innovations, and challenges in the technology sphere. We also welcome discussions on the intersections of technology and society. If it’s technological news or discussion of technology, it probably belongs here.

Remember the overriding ethos on Beehaw: Be(e) Nice. Each user you encounter here is a person, and should be treated with kindness (even if they’re wrong, or use a Linux distro you don’t like). Personal attacks will not be tolerated.

Subcommunities on Beehaw:


This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.

founded 2 years ago
MODERATORS
 

I'm new to the field of large language models (LLMs) and I'm really interested in learning how to train and use my own models for qualitative analysis. However, I'm not sure where to start or what resources would be most helpful for a complete beginner. Could anyone provide some guidance and advice on the best way to get started with LLM training and usage? Specifically, I'd appreciate insights on learning resources or tutorials, tips on preparing datasets, common pitfalls or challenges, and any other general advice or words of wisdom for someone just embarking on this journey.

Thanks!

you are viewing a single comment's thread
view the rest of the comments
[–] [email protected] 3 points 7 months ago (3 children)

OLlama is so fucking slow. Even with a 16-core overclocked Intel on 64Gb RAM with an Nvidia 3080 10Gb VRAM, using a 22B parameter model, the token generation for a simple haiku takes 20 minutes.

[–] xcjs 5 points 7 months ago (1 children)

No offense intended, but are you sure it's using your GPU? Twenty minutes is about how long my CPU-locked instance takes to run some 70B parameter models.

On my RTX 3060, I generally get responses in seconds.

[–] [email protected] 3 points 7 months ago

I agree. My 3070 runs the 8B Llama3 model in about 250ms, especially for short responses.

[–] xcjs 1 points 7 months ago (1 children)

Ok, so using my "older" 2070 Super, I was able to get a response from a 70B parameter model in 9-12 minutes. (Llama 3 in this case.)

I'm fairly certain that you're using your CPU or having another issue. Would you like to try and debug your configuration together?

[–] [email protected] 1 points 7 months ago (1 children)

I think I fucked up my docker setup and will wipe and start over.

[–] xcjs 1 points 7 months ago (1 children)

Good luck! I'm definitely willing to spend a few minutes offering advice/double checking some configuration settings if things go awry again. Let me know how things go. :-)

[–] [email protected] 2 points 7 months ago (1 children)

My setup is Win 11 Pro ➡️ WSL2 / Debian ➡️ Docker Desktop (for windows)

Should I install the nvidia drivers within Debian even though the host OS already has drivers?

[–] xcjs 1 points 7 months ago* (last edited 7 months ago) (1 children)

I think there was a special process to get Nvidia working in WSL. Let me check... (I'm running natively on Linux, so my experience doing it with WSL is limited.)

https://docs.nvidia.com/cuda/wsl-user-guide/index.html - I'm sure you've followed this already, but according to this, it looks like you don't want to install the Nvidia drivers, and only want to install the cuda-toolkit metapackage. I'd follow the instructions from that link closely.

You may also run into performance issues within WSL due to the virtual machine overhead.

[–] [email protected] 2 points 7 months ago (1 children)

I did indeed follow that guide already, thank you for the respect; I am an idiot and installed both the nvidia WSL driver on top of the host OS driver _as well as the Cuda driver. So I'll try again with only that guide and see what breaks.

[–] xcjs 1 points 7 months ago

We all mess up! I hope that helps - let me know if you see improvements!

[–] [email protected] 1 points 7 months ago* (last edited 7 months ago) (2 children)

Hmmm weird. I have a 4090 / Ryzen 5800X3D and 64GB and it runs really well. Admittedly it's the 8B model because the intermediate sizes aren't out yet and 70B simply won't fly on a single GPU.

But it really screams. Much faster than I can read. PS: Ollama is just llama.cpp under the hood.

Edit: Ah, wait, I know what's going wrong here. The 22B parameter model is probably too big for your VRAM. Then it gets extremely slow yes.

[–] xcjs 1 points 7 months ago* (last edited 7 months ago)

It should be split between VRAM and regular RAM, at least if it's a GGUF model. Maybe it's not, and that's what's wrong?

[–] [email protected] 1 points 7 months ago (1 children)

What is the appropriate size for 10Gb VRAM?

[–] [email protected] 2 points 7 months ago

It depends on your prompt/context size too. The more you have the more memory you need. Try to check the memory usage of your GPU with GPU-Z with different models and scenarios.