MTP weights?

#45

by SzymonOzog - opened Mar 11

Mar 11

Does this contain multi token prediciton weights? I know that they are marked as layer 61 in the huggingface implementation but I can see this goes up to blk_60. Are they named differently or are they just prunned?

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