fix incorrect attention head dimension in AttnProcessor2_0 (huggingface#4154)

fix inner_dim

Author: zhvng

src/diffusers/models/attention_processor.py

@@ -1096,7 +1096,6 @@ def __call__(
         batch_size, sequence_length, _ = (
             hidden_states.shape if encoder_hidden_states is None else encoder_hidden_states.shape
         )
-        inner_dim = hidden_states.shape[-1]
 
         if attention_mask is not None:
             attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length, batch_size)
@@ -1117,6 +1116,7 @@ def __call__(
         key = attn.to_k(encoder_hidden_states)
         value = attn.to_v(encoder_hidden_states)
 
+        inner_dim = key.shape[-1]
         head_dim = inner_dim // attn.heads
 
         query = query.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)