Update app.py

app.py

@@ -20,7 +20,7 @@ MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
 if torch.cuda.is_available():
     model_id = "BramVanroy/GEITje-7B-ultra-sft"
-    model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto")
+    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
     tokenizer = AutoTokenizer.from_pretrained(model_id)