fix fuyu device_map compatibility (#29880)

fix foward

src/transformers/models/fuyu/modeling_fuyu.py

@@ -290,7 +290,9 @@ class FuyuForCausalLM(FuyuPreTrainedModel):
             inputs_embeds = self.language_model.get_input_embeddings()(input_ids)
             if image_patches is not None and past_key_values is None:
                 patch_embeddings = [
-                    self.vision_embed_tokens(patch.to(self.vision_embed_tokens.weight.dtype)).squeeze(0)
+                    self.vision_embed_tokens(patch.to(self.vision_embed_tokens.weight.dtype))
+                    .squeeze(0)
+                    .to(inputs_embeds.device)
                     for patch in image_patches
                 ]
                 inputs_embeds = self.gather_continuous_embeddings(