AttentionProcessor.group_norm num_channels should be query_dim

The group_norm on the attention processor should really norm the number
of channels in the query _not_ the inner dim. This wasn't caught before
because the group_norm is only used by the added kv attention processors
and the added kv attention processors are only used by the karlo models
which are configured such that the inner dim is the same as the query
dim.

Author: williamberman

src/diffusers/models/attention_processor.py

@@ -81,7 +81,7 @@ def __init__(
         self.added_kv_proj_dim = added_kv_proj_dim
 
         if norm_num_groups is not None:
-            self.group_norm = nn.GroupNorm(num_channels=inner_dim, num_groups=norm_num_groups, eps=1e-5, affine=True)
+            self.group_norm = nn.GroupNorm(num_channels=query_dim, num_groups=norm_num_groups, eps=1e-5, affine=True)
         else:
             self.group_norm = None