LocalLLaMA

2929 readers

23 users here now

Welcome to LocalLLaMA! Here we discuss running and developing machine learning models at home. Lets explore cutting edge open source neural network technology together.

Get support from the community! Ask questions, share prompts, discuss benchmarks, get hyped at the latest and greatest model releases! Enjoy talking about our awesome hobby.

As ambassadors of the self-hosting machine learning community, we strive to support each other and share our enthusiasm in a positive constructive way.

founded 2 years ago

MODERATORS

[email protected]

Training a model without a GPU (sh.itjust.works)

submitted 1 year ago by [email protected] to c/[email protected]

9 comments fedilink hide all child comments

Hi, I'm currently starting to learn how LLM works in depth, so I started using nanoGPT to understand how to train a model and I'd like to play around with the code a little more. So I set myself a goal to train a model that can write basic French, it doesn't to be coherent or deep in its writing, just French with correct grammar. I only have a laptop that doesn't have a proper GPU, so I can't really train a model with billions of parameters. Do you think it's possible without too much dataset or intensive training? Is it a better idea if I use something different from nanoGPT?

TLDR: I'd like to train my own LLM on my laptop which doesn't have a GPU. It's only for learning purpose, so my goal is that it can write basic French. Is it doable? If it is, do you have any tips to make this easier?

you are viewing a single comment's thread
view the rest of the comments

[–] [email protected] 4 points 1 year ago* (last edited 1 year ago) (2 children)

TL;DR yeah, it's doable, just slow.

You can train without a GPU, it just takes longer. More RAM and a better CPU will help up to a point. I don't think text generation is a particularly difficult task- you could probably do it with something like a Markov chain rather than an LLM if you don't care whether it's particularly coherent.

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

Well, I use my laptop as a daily-driver, so training an AI in the background, even when I don't use it seems a bit complicated. The Markov chain seems like an interesting alternative for what I'm looking, does any tools to use one exist or should I build one from scratch?

[–] [email protected] 2 points 1 year ago

There are libraries that can do it. Here's one: https://pypi.org/project/PyDTMC/