Update app.py

app.py

@@ -25,7 +25,8 @@ tokenizer = AutoTokenizer.from_pretrained(model_id)
 model = AutoModelForCausalLM.from_pretrained(model_id,
                                              # load_in_8bit=True,
                                              quantization_config=nf4_config,
-                                             torch_dtype = torch.bfloat16,
+                                            attn_implementation="flash_attention_2",
+                                             # torch_dtype = torch.bfloat16,
                                              device_map="auto"
                                             )