fix from_vision_text_pretrained doc example (#15453)

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

src/transformers/models/vision_text_dual_encoder/modeling_flax_vision_text_dual_encoder.py

@@ -449,7 +449,7 @@ class FlaxVisionTextDualEncoderModel(FlaxPreTrainedModel):
 
         >>> # initialize a model from pretrained ViT and BERT models. Note that the projection layers will be randomly initialized.
         >>> model = FlaxVisionTextDualEncoderModel.from_vision_text_pretrained(
-        ...     "bert-base-uncased", "google/vit-base-patch16-224"
+        ...     "google/vit-base-patch16-224", "bert-base-uncased"
         ... )
         >>> # saving model after fine-tuning
         >>> model.save_pretrained("./vit-bert")

src/transformers/models/vision_text_dual_encoder/modeling_vision_text_dual_encoder.py

@@ -469,7 +469,7 @@ class VisionTextDualEncoderModel(PreTrainedModel):
 
         >>> # initialize a model from pretrained ViT and BERT models. Note that the projection layers will be randomly initialized.
         >>> model = VisionTextDualEncoderModel.from_vision_text_pretrained(
-        ...     "bert-base-uncased", "google/vit-base-patch16-224"
+        ...     "google/vit-base-patch16-224", "bert-base-uncased"
         ... )
         >>> # saving model after fine-tuning
         >>> model.save_pretrained("./vit-bert")