Update readme

Signed-off-by: Jael Gu <mengjia.gu@zilliz.com>

README.md

@@ -352,14 +352,6 @@ and then return text embedding in ndarray.
 If data is string, the operator returns embedding(s) in ndarray.
 If data is a list, the operator returns embedding(s) in a list.
 
-
-**Returns**:
-
-*numpy.ndarray*
-
-​	The text embedding extracted by model.
-
-
 <br />
 
 ***save_model(format='pytorch', path='default')***
@@ -408,6 +400,16 @@ print(f'Onnx-support/Total Models: {len(onnx_list)}/{len(full_list)}')
     2022-12-13 16:25:15,916 - 140704500614336 - auto_transformers.py-auto_transformers:68 - WARNING: The operator is initialized without specified model.
     Onnx-support/Total Models: 111/126
 
+<br />
+
+**Returns**:
+
+*numpy.ndarray or list*
+
+​	The text embedding (or token embeddings) extracted by model.
+
+<br />
+
 ## Fine-tune
 ### Requirement
 If you want to train this operator, besides dependency in requirements.txt, you need install these dependencies.