Open raywanb opened 2 months ago
Awesome, very excited for VLLM support! The control vector gets added after the MLP/FC layer. However, I noticed you linked to Mixtral, not sure if that was intentional but FYI (I should note this more clearly in the README) control vectors don't work on MoEs currently, probably best to implement for dense models only.
Ok, got it thanks. btw is there a proper way of saving control vectors so ppl can upload them onto HuggingFace?
Ok, got it thanks. btw is there a proper way of saving control vectors so ppl can upload them onto HuggingFace?
This might help! https://github.com/vgel/repeng/issues/23
@raywanb
Ok, got it thanks. btw is there a proper way of saving control vectors so ppl can upload them onto HuggingFace?
There's an official way now, see #34 !
I'm trying to implement control vector into vllm codebase for mixtral model, but I was wondering where should I add the control vector to the layer. Should it be added before attention, fully connected, or after? Thanks @vgel
https://github.com/vllm-project/vllm/blob/a53222544c6385ee314a26fdf42eb14f5b4e5ad9/vllm/model_executor/models/mixtral.py#L270