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
[Bugfix] update the prefix for qwen2 (#11795)
Co-authored-by: jiadi.jjd <jiadi.jjd@antgroup.com>
jiangjiadi committed
Jan 8, 2025 at 02:36 UTC
c994223d569221652643e897d8402b835ead411d
1 file changed
+1
-1
vllm/model_executor/models/qwen2.py
+1
-1
@@ -298,7 +298,7 @@ class Qwen2Model(nn.Module):
298
lambda prefix: Qwen2DecoderLayer(config=config,
299
cache_config=cache_config,
300
quant_config=quant_config,
301
- prefix=f"{prefix}.layers"),
301
+ prefix=prefix),
302
prefix=f"{prefix}.layers",
303
)
304