Update README.md

README.md

@@ -63,7 +63,7 @@ def get_response(question, context, max_length=32):
                attention_mask=features['attention_mask'].to(device),
                max_length=max_length)
 
-  return tokenizer.decode(output[0])
+  return tokenizer.decode(output[0], skip_special_tokens=True)
   
 # Some examples in different languages