Update app.py

app.py

@@ -112,7 +112,7 @@ def main(query: str, client: QdrantClient, collection_name: str, llm: Llama, den
     response = llm.generate(prompts=inputs_1, sampling_params=sampling_params_1)
 
     text = response[0].outputs[0].text
-    print(f'TEXT: {text}')
+    print(f'TEXT: {response}')
 
     sampling_params_2 = vllm.SamplingParams(temperature=0.75, max_tokens=3000, stop_token_ids=stop_token_ids)
     prompt_2 = [
@@ -420,8 +420,7 @@ if __name__ == '__main__':
             if st.session_state.chat_id not in conversations.keys():
                 stop_token_ids = [151329, 151336, 151338]
                 sampling_params = vllm.SamplingParams(temperature=0.75, max_tokens=35, stop_token_ids=stop_token_ids)
-                str = f"{prompt}\nExplain the above in one sentence:"
-                prompt = [{"role": "user", "content": str}]
+                prompt = [{"role": "user", "content": f"{prompt}\nExplain the above in one sentence:"}]
                 inputs = tokenizer.apply_chat_template(prompt, tokenize=False, add_generation_prompt=True)
                 outputs = llm.generate(prompts=inputs, sampling_params=sampling_params)