Update README.md

README.md

@@ -14,6 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
 # how to use the model
 
 # Load model directly
+
+```
 `from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
 import torch
 
@@ -28,6 +30,7 @@ with torch.no_grad():
     generated_text = tokenizer.decode(outputs[0])
 
 print(generated_text)`
+```
 
 # checkpoint