Update app.py

app.py

@@ -389,7 +389,7 @@ else:
 #     )quantization_config = nf4_config, 
 
 tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b-it", token=HF_TOKEN)
-model = AutoModelForCausalLM.from_pretrained("google/gemma-2b-it", device_map="auto", load_in_4bit=True,token=HF_TOKEN)
+model = AutoModelForCausalLM.from_pretrained("google/gemma-2b-it", token=HF_TOKEN)
 
 
 def make_inference(query, hist):