Small DeBERTa Docstring Fixes (#666)

* Small DeBERTa docstring fixes

* Small edit

Author: abheesht17

keras_nlp/models/deberta_v3/deberta_v3_presets.py

@@ -90,7 +90,7 @@
         },
         "preprocessor_config": {},
         "description": (
-            "Base size of DeBERTaV3. "
+            "Large size of DeBERTaV3. "
             "Trained on English Wikipedia, BookCorpus and OpenWebText."
         ),
         "weights_url": "https://storage.googleapis.com/keras-nlp/models/deberta_v3_large_en/v1/model.h5",

keras_nlp/models/deberta_v3/deberta_v3_tokenizer.py

@@ -97,7 +97,7 @@ def from_preset(
         preset,
         **kwargs,
     ):
-        """Instantiate anDeBERTa tokenizer from preset vocabulary.
+        """Instantiate a DeBERTa tokenizer from preset vocabulary.
 
         Args:
             preset: string. Must be one of {{names}}.