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
[Misc] Unified linear print info (#23516)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Jee Jee Li committed
Aug 25, 2025 at 11:13 UTC
170e8ea9ea95294d1bdc4af4bea73741ea759e22
1 file changed
+1
-1
vllm/model_executor/layers/linear.py
+1
-1
@@ -1378,7 +1378,7 @@ class RowParallelLinear(LinearBase):
1378
return output, output_bias
1379
1380
def extra_repr(self) -> str:
1381
- s = f"input_features={self.input_size_per_partition}"
1381
+ s = f"in_features={self.input_size_per_partition}"
1382
s += f", output_features={self.output_size}"
1383
s += f", bias={self.bias is not None}"
1384
s += f", tp_size={self.tp_size}"