How can I combine the MLP projector weights with Vicuna-7B-v1.5 (provided in the Files and versions) to obtain a base vision-language model (VLM) for evaluation, and are there any scripts available to integrate the projector with the base LLM?
· Sign up or log in to comment
