update model card

README.md

@@ -12,11 +12,11 @@ You can also fine-tune this model on similar down-stream tasks, such as a dialog
 ## How to use 
 You can use this model directly with:
 
-'''
-from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
-tokenizer = AutoTokenizer.from_pretrained("LiqiangXiao/ConvSearch_QU")
-model = AutoModelForSeq2SeqLM.from_pretrained("LiqiangXiao/ConvSearch_QU")
-'''
+  
+    from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+    tokenizer = AutoTokenizer.from_pretrained("LiqiangXiao/ConvSearch_QU")
+    model = AutoModelForSeq2SeqLM.from_pretrained("LiqiangXiao/ConvSearch_QU")
+
 
 ## Training data
 ConvSearch was pretrained on a dialog corpus with 49,999 dialogs/942,766 turns.