Update README.md

README.md

@@ -37,7 +37,10 @@ This model is electra-small finetuned for NER prediction task. The model current
 
 ## Uses
 
-This model fine tuning is like sub - word token prediciton. So the model may predict wrong entities for same word (different sub word). Use 'aggregation_strategy' to "max" when using transformer's pipeline.
+This model uses tokenizer that is from distilbert family. So the model may predict wrong entities for same word (different sub word). Use 'aggregation_strategy' to "max" when using transformer's pipeline.
+for example 'ashwin ::"
+    ash" => Person 
+    win  => Location
 
 
 ### Out-of-Scope Use