Apply suggestions from code review

Co-authored-by: Mohit Sharma <[email protected]>
huggingface · Jan 30, 2025 · f56e24b · f56e24b
1 parent 9376066
commit f56e24b
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 4 deletions.
diff --git a/Dockerfile_amd b/Dockerfile_amd
@@ -279,9 +279,9 @@ RUN git clone https://github.com/danieldk/marlin-kernels.git && \
 
 FROM kernel-builder AS moe-kernels
 WORKDIR /usr/src
-ENV MOE_KERNELS_BRANCH=9fb72ccb00b8da96a41f9b9c48f305c7dbd266bc
+ENV MOE_KERNELS_BRANCH=d7e042bf9f7aff10c631212fc71b24895d66eb59
 ENV VLLM_TARGET_DEVICE=rocm
-RUN git clone https://github.com/mht-sharma/moe-kernels.git && \
+RUN git clone https://github.com/danieldk/moe-kernels.git && \
     cd moe-kernels && \
     git checkout ${MOE_KERNELS_BRANCH} && \
     python setup.py install

diff --git a/server/text_generation_server/layers/moe/unquantized.py b/server/text_generation_server/layers/moe/unquantized.py
@@ -73,8 +73,6 @@ def forward(self, x: torch.Tensor, *, gating_output: torch.Tensor) -> torch.Tens
                 num_expert_group=self.n_expert_group,
                 topk_group=self.topk_group,
             )
-        # from loguru import logger
-        # logger.info("Fused MoE is used here")
         return fused_moe(
             x,
             w1=self.gate_up_proj,