Update app.py

app.py

@@ -20,7 +20,7 @@ quant_config = BitsAndBytesConfig(
     bnb_4bit_use_double_quant=True,
     bnb_4bit_compute_dtype=bfloat16
 )
-tokenizer = LlamaTokenizer.from_pretrained(model_name)
+tokenizer = LlamaTokenizer.from_pretrained(model_name, use_fast=False)
 model = LlamaForCausalLM.from_pretrained(
     model_name,
     quantization_config=quant_config