Refactor CogVideoX transformer forward (huggingface#10789)

update

Author: a-r-r-o-w

src/diffusers/models/transformers/cogvideox_transformer_3d.py

@@ -503,14 +503,7 @@ def forward(
                     attention_kwargs=attention_kwargs,
                 )
 
-        if not self.config.use_rotary_positional_embeddings:
-            # CogVideoX-2B
-            hidden_states = self.norm_final(hidden_states)
-        else:
-            # CogVideoX-5B
-            hidden_states = torch.cat([encoder_hidden_states, hidden_states], dim=1)
-            hidden_states = self.norm_final(hidden_states)
-            hidden_states = hidden_states[:, text_seq_length:]
+        hidden_states = self.norm_final(hidden_states)
 
         # 4. Final block
         hidden_states = self.norm_out(hidden_states, temb=emb)