search

openlm-research / open_llama

OpenLLaMA, a permissively licensed open source reproduction of Meta AI’s LLaMA 7B trained on the RedPajama dataset
Apache License 2.0
7.36k stars 374 forks source link

Thanks! Does it fit in Free Colab? #1

Closed amrrs closed 1 year ago

amrrs commented 1 year ago

Hey Thanks very much for the effort. Looking at the released model size, Wondering if it can fit in Free Colab?

young-geng commented 1 year ago

The model architecture and size is the same as LLaMA 7B, so if the original LLaMA 7B fits, our model should fit as well. If not, we are also working on a smaller LLaMA 3B model, which we will release a preview checkpoint soon.

swyxio commented 1 year ago

@amrrs if you manage to get it working in a Colab i'd be very keen too!

curiousily commented 1 year ago

I made a video on how to run it in Google Colab (it fits in T4 GPU): https://www.youtube.com/watch?v=1NOPciKuQb8

Hope that helps!

young-geng commented 1 year ago

We've just released a checkpoint for our 3B model and that should definitely fit in colab.