Update app.py

app.py

@@ -42,7 +42,10 @@ model = AutoPeftModelForCausalLM.from_pretrained(
         model_id, # YOUR MODEL YOU USED FOR TRAINING
         load_in_4bit = load_in_4bit,
     )
-tokenizer = AutoTokenizer.from_pretrained(model_id)
+tokenizer = AutoTokenizer.from_pretrained(
+            model_id,
+            quantization_config=BitsAndBytesConfig(load_in_8bit=True)
+    )
 
 model.config.sliding_window = 4096
 model.eval()