@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 46fa98ccad
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 03:15 UTC

[Misc] Clean up debug code in Deepseek-V3 (#11930)

Signed-off-by: Isotr0py <2037008807@qq.com>

Isotr0py committed Jan 11, 2025 at 03:19 UTC 46fa98ccad444dbacc7f95995b79f65ddab3ff7c
1 file changed -3
vllm/model_executor/models/deepseek_v3.py
-3
@@ -639,9 +639,6 @@ class DeepseekV3ForCausalLM(nn.Module, SupportsPP):
639 if is_pp_missing_parameter(name, self):
640 continue
641
642 - if name not in params_dict:
643 - for key in params_dict:
644 - print(key)
642 param = params_dict[name]
643 weight_loader = getattr(param, "weight_loader",
644 default_weight_loader)