Advice - Getting started with LLMs (beehaw.org)

submitted 1 week ago by its_me_xiphos@beehaw.org to c/technology@beehaw.org

23 comments fedilink hide all child comments

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!

all 24 comments

sorted by: hot top controversial new old

[-] Zworf@beehaw.org 17 points 1 week ago* (last edited 1 week ago)

Training your own will be very difficult. You will need to gather so much data to get a model that has basic language understanding.

What I would do (and am doing) is just taking something like llama3 or mistral and adding your own content using RAG techniques.

But fair play if you do manage to train a real model!

[-] BaroqueInMind@lemmy.one 3 points 1 week ago

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@programming.dev 5 points 1 week ago

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.

[-] kiku123@feddit.de 3 points 1 week ago

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

[-] xcjs@programming.dev 1 points 1 week ago

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?

[-] BaroqueInMind@lemmy.one 1 points 1 week ago

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

[-] xcjs@programming.dev 1 points 1 week ago

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. :-)

[-] BaroqueInMind@lemmy.one 2 points 1 week ago

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@programming.dev 1 points 1 week ago* (last edited 1 week ago)

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.

[-] BaroqueInMind@lemmy.one 2 points 1 week ago

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@programming.dev 1 points 1 week ago

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

[-] Zworf@beehaw.org 1 points 1 week ago* (last edited 1 week ago)

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.

[-] BaroqueInMind@lemmy.one 1 points 1 week ago

What is the appropriate size for 10Gb VRAM?

[-] Zworf@beehaw.org 2 points 6 days 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.

[-] xcjs@programming.dev 1 points 1 week ago* (last edited 1 week 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?

[-] makingStuffForFun@lemmy.ml 5 points 1 week ago

I'm also interested, so I hope you don't mind me joining the ride. Personally, I'd like a self hosted tool, but am happy to see what the community says.

[-] trevron@beehaw.org 4 points 1 week ago

If you just want to use a local llm, using something like gpt4all is probably the easiest. Oobabooga or llama.cpp for a more advanced route.

I use ollama with llama3 on my macbook with open-webui and it works real nice. Mistral7b is another one I like. On my PC I have been using oobabooga with models I get from huggingface and I use it as an api for hobby projects.

I have never trained models, I don't have the vram. My GPU is pretty old so I just use these for random gamedev and webdev projects and for messing around with RP in sillytavern.

[-] TehPers@beehaw.org 3 points 1 week ago

I managed to get ollama running through Docker easily. It's by far the least painful of the options I tried, and I just make requests to the API it exposes. You can also give it GPU resources through Docker if you want to, and there's a CLI tool for a quick chat interface if you want to play with that. I can get LLAMA 3 (8B) running on my 3070 without issues.

Training a LLM is very difficult and expensive. I don't think it's a good place for anyone to start. Many of the popular models (LLAMA, GPT, etc) are astronomically expensive to train and require and ungodly number of resources.

[-] trevron@beehaw.org 1 points 1 week ago

yep, definitely agree with all of this.

[-] Midnitte@beehaw.org 2 points 1 week ago

Using LM Studio would be even easier to get started

[-] xcjs@programming.dev 2 points 1 week ago

Unfortunately, I don't expect it to remain free forever.

[-] its_me_xiphos@beehaw.org 1 points 1 week ago

I really appreciate all the responses, but I'm overwhelmed by the amount of information and possible starting points. Could I ask you to explain or reference learning content that talks to me like I'm a curious five year old?

ELI 5?