Update app.py

app.py

@@ -15,7 +15,7 @@ tuned_adapter = "newronai/llama-2-7b-QLoRA-Trial1"
 config = PeftConfig.from_pretrained(tuned_adapter)
 model = AutoModelForCausalLM.from_pretrained(base_model,
                                              use_cache="cache",
-                                             
+                                             load_in_8bit = True
                                              # quantization_config=bnb_config
                                             )