transformers
[Gemma Quant] Preserve Gemma 4 MoE router projection in full precision during quantization
#48801
Open
Open
[Gemma Quant] Preserve Gemma 4 MoE router projection in full precision during quantization #48801
pculliton
wants to merge
1 commit into
huggingface:main
from
pculliton:gemma-quant-router-fix
Login to write a write a comment.
Login via GitHub