meta-llama

/

Meta-Llama-3-70B-Instruct

like 1.05k

Hi, I'm working on llama 3 support for gpt-fast.

And have a working version, However it requires to save the model in the pytorch_model.bin format. Could you upload that version pls?

I have a copy here, but it would be useful if the files are in another folder of something in the official repo.
https://huggingface.co/eastwind/llama-3-70b-instruct-hf-pt/tree/main

Thanks .

Gpt-fast PR.
https://github.com/pytorch-labs/gpt-fast/pull/169