xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-06-02 17:37:56 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix typo

Browse Source 
Signed-off-by: yuantao <2422264527@qq.com>
This commit is contained in:
yuantao 2025-11-26 14:40:57 +08:00
parent ca7eaa045c
commit 0abdaad39b
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/v1/worker/gpu_model_runner.py
View File

@ -5319,7 +5319,7 @@ class GPUModelRunner(
return module return module
raise KeyError(f"Module '{target_name}' not found") raise KeyError(f"Module '{target_name}' not found")
for layer_name, kv_cache in kv_caches.item(): for layer_name, kv_cache in kv_caches.items():
layer_prefix = layer_name.rsplit(".", 1)[0] layer_prefix = layer_name.rsplit(".", 1)[0]
self_attn_module = find_module_by_name(self.model, layer_prefix) self_attn_module = find_module_by_name(self.model, layer_prefix)
if not hasattr(self_attn_module, "get_sink_kv"): if not hasattr(self_attn_module, "get_sink_kv"):