fixing error

app.py

@@ -91,6 +91,14 @@ model = AutoModelForCausalLM.from_pretrained(
     cache_dir=CACHE_DIR
 )
 
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    model_kwargs={"torch_dtype": torch.bfloat16},
+    # device="cuda",
+)
+
 def format_prompt(prompt, retrieved_documents, k):
     """using the retrieved documents we will prompt the model to generate our responses"""
     PROMPT = f"Pertanyaan:{prompt}\nKonteks:"
@@ -104,10 +112,11 @@ def chat_function(message, history, max_new_tokens=256, temperature=0.6):
     
     messages = [{"role":"system","content":SYS_MSG},
                 {"role":"user", "content":formatted_prompt}]
-    prompt = pipeline.tokenizer.apply_chat_template(
-        messages,
-        tokenize=False,
-        add_generation_prompt=True,)
+    # prompt = pipeline.tokenizer.apply_chat_template(
+    #     messages,
+    #     tokenize=False,
+    #     add_generation_prompt=True,)
+    prompt = tokenizer(formatted_prompt, return_tensors="pt")
     print(f"Prompt: {prompt}\n")
     terminators = [
         pipeline.tokenizer.eos_token_id,