KeyError: 'embed_tokens' when Running vLLM

#2
by CedPei - opened

Hello,
I'm currently working on integrating the model with vLLM and have encountered an issue during the loading process. When attempting to load the model, I receive the following error:

param = params_dict[name]
KeyError: 'embed_tokens'

Any insights or suggestions from the community would be greatly appreciated.
Thank you for your support.

Sign up or log in to comment