Update utils.py

utils.py

@@ -376,7 +376,7 @@ def query(api_llm, payload):
 
 def llm_chain2(prompt, context):
     full_prompt = RAG_CHAIN_PROMPT.format(context=context, question=prompt)
-    inputs = tokenizer_summarization(full_prompt, return_tensors="pt", max_length=1024, truncation=True)
+    inputs = tokenizer_rag(full_prompt, return_tensors="pt", max_length=1024, truncation=True)
 
     #Generiere die Antwort
     outputs = modell_rag.generate(inputs['input_ids'], max_length=1024, num_beams=2, early_stopping=True)