Update utils.py

utils.py

@@ -396,7 +396,7 @@ def llm_chain2(prompt, context):
         max_new_tokens=1024, 
         do_sample=True, 
         temperature=0.9, 
-        pad_token_id=tokenizer.eos_token_id
+        pad_token_id=tokenizer_rag.eos_token_id
     )
     #outputs = modell_rag.generate(inputs['input_ids'], max_new_tokens=1024, num_beams=2, early_stopping=True)
     answer = tokenizer_rag.decode(outputs[0], skip_special_tokens=True)