Update app.py

app.py

@@ -11,6 +11,7 @@ llm = Llama(model_path= hf_hub_download(repo_id="TheBloke/airoboros-l2-13b-gpt4-
 history = ""
 
 def generate_text(input_text, history):
+    history  = history[-1]
     full_conversation = f"{history} Q: {input_text} \n A:"
     output = llm(full_conversation, max_tokens=1024, stop=["Q:", "\n"], echo=True)
     response = output['choices'][0]['text']