Request for model definition

by AvDy - opened 9 days ago

Discussion

AvDy

9 days ago

•

edited 9 days ago

Hi,

Thank you for your great work. If possible, I request you to also provide the model definition for this pretrained model. As mentioned, it is using a LLaMA3 8B model, but I assume that some of the functions have been changed for the quantization purposes.

medmekk

Hugging Face 1Bit LLMs org 8 days ago

Hello, The model definition is the same as a Llama3 8B, the only difference is Linear layers are replaced with BitLinear layers. You can refer to the blogpost here to understand how BitLinear works

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment