Fix blender bot missleading doc (#20301)

* fix the doc to specify that add_prefix_space = False

* add correct expected output

src/transformers/models/blenderbot/tokenization_blenderbot.py

@@ -99,8 +99,9 @@ class BlenderbotTokenizer(PreTrainedTokenizer):
     ```
     >>> from transformers import BlenderbotTokenizer
     >>> tokenizer = BlenderbotTokenizer.from_pretrained("facebook/blenderbot-3B")
+    >>> tokenizer.add_prefix_space = False
     >>> tokenizer("Hello world")['input_ids']
-    [6950, 1085, 2]
+    [47, 921, 86, 1085, 2]
     >>> tokenizer(" Hello world")['input_ids']
     [6950, 1085, 2]
     ```