Update app.py

app.py

@@ -103,7 +103,7 @@ def submit_message(prompt, prompt_template, temperature, max_tokens, context_len
 # completion = openai.ChatCompletion.create(model="gpt-3.5-turbo", messages=system_prompt + history[-context_length*2:] + [prompt_msg], temperature=temperature, max_tokens=max_tokens)
 
         completion = RetrievalQA.from_chain_type(llm=ChatOpenAI(temperature=temperature, max_tokens=max_tokens, model_name="gpt-3.5-turbo"), chain_type="stuff", retriever=vectordb.as_retriever() , return_source_documents=True)
-        completion = completion({"query": system_prompt + history[-context_length*2:] + [prompt_msg] })
+        completion = completion({"query": system_prompt +  [prompt_msg] })
         # from https://blog.devgenius.io/chat-with-document-s-using-openai-chatgpt-api-and-text-embedding-6a0ce3dc8bc8
         
         history.append(prompt_msg)
@@ -118,10 +118,10 @@ def submit_message(prompt, prompt_template, temperature, max_tokens, context_len
             "content": f"Error: {e}"
         })
 
-    #total_tokens_used_msg = f"Total tokens used: {state['total_tokens']}"
+    total_tokens_used_msg = f"Total tokens used: {state['total_tokens']}"
     chat_messages = [(history[i]['content'], history[i+1]['content']) for i in range(0, len(history)-1, 2)]
 
-    return '', chat_messages #, total_tokens_used_msg, state
+    return '', chat_messages, total_tokens_used_msg, state
 
 def clear_conversation():
     return gr.update(value=None, visible=True), None, "", get_empty_state()