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 14, 05:30 UTC
[easy] fix pre commit error on trunk (#26665)
Signed-off-by: Huamin Li <3ericli@gmail.com>
Huamin Li committed
Oct 12, 2025 at 14:25 UTC
4fca1a1bd25a1b0d3b49f3fa832425cef5a612fb
1 file changed
+2
-2
vllm/model_executor/models/bert.py
+2
-2
@@ -609,8 +609,8 @@ class SPLADESparsePooler(Pooler):
609
def __init__(
610
self,
611
mlm_head: nn.Module,
612
- cls_token_id: Optional[int] = 101,
613
- sep_token_id: Optional[int] = 102,
612
+ cls_token_id: int | None = 101,
613
+ sep_token_id: int | None = 102,
614
pooling: str = "max",
615
remove_cls_sep: bool = True,
616
):