Skip to content

Commit b6636c8

Browse files
committed
fix typo
Signed-off-by: Kyle Sayers <[email protected]>
1 parent e406a2e commit b6636c8

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

src/compressed_tensors/modeling/attention.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -133,7 +133,7 @@ def register_query_hook(module: torch.nn.Module, hook: Callable) -> RemovableHan
133133
impl = getattr(module, IMPL_ATTR)
134134

135135
def _hook(impl: QuantizedAttentionImpl, args, kwargs):
136-
bound = inspect.signature(module.forward).bind(*args, **kwargs)
136+
bound = inspect.signature(impl.forward).bind(*args, **kwargs)
137137
value = hook(module, bound.arguments["query"])
138138
if value is not None:
139139
bound.arguments["query"] = value

0 commit comments

Comments
 (0)