[Bug] Fix usage of .transpose() and .view() consecutively. (#11979)

Author: liaoyanqing666

vllm/attention/layer.py

@@ -230,7 +230,7 @@ def forward(
                                                  value,
                                                  scale=self.scale)
             out = out.transpose(1, 2)
-        return out.view(bsz, q_len, -1)
+        return out.reshape(bsz, q_len, -1)
 
 
 def unified_attention(

vllm/model_executor/models/intern_vit.py

@@ -271,7 +271,7 @@ def forward(self, x: torch.Tensor) -> torch.Tensor:
         v = v.transpose(1, 2)
 
         x = F.scaled_dot_product_attention(q, k, v, scale=self.scale)
-        x = x.transpose(1, 2).view(B, N, -1)
+        x = x.transpose(1, 2).reshape(B, N, -1)
 
         x = self.proj(x)
         return x