Update app.py

app.py

@@ -12,6 +12,11 @@ def load_model_and_tokenizer():
         revision="main"
     )
     tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=True)
+    
+    # Check if tokenizer has a pad token, if not add it
+    if tokenizer.pad_token is None:
+        tokenizer.pad_token = tokenizer.eos_token  # Use eos_token as padding token
+    
     return model, tokenizer
 
 model, tokenizer = load_model_and_tokenizer()
@@ -33,7 +38,7 @@ def get_response(comment):
         input_ids=inputs["input_ids"].to("cuda"), 
         attention_mask=inputs["attention_mask"].to("cuda"), 
         max_new_tokens=140, 
-        pad_token_id=tokenizer.eos_token_id
+        pad_token_id=tokenizer.pad_token_id  # Ensure padding is handled properly
     )
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)
     return response.split("[/INST]")[-1].strip()