From your work, I find a new way to do model ensemble

#14

by xxx1 - opened Mar 26

xxx1

Mar 26

num_local_experts== num_experts_per_tok means all experts is in use. But diff rate of diff model

Owner Mar 26

so any new model we can use?

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