Update app.py

app.py

@@ -45,7 +45,7 @@ model = AutoModelForCausalLM.from_pretrained(model_id, token= token,
                                                  # torch_dtype= torch.uint8, 
                                              torch_dtype=torch.float16,
                                               # torch_dtype=torch.fl,
-                                             # attn_implementation="flash_attention_2",
+                                             attn_implementation="flash_attention_2",
                                              # low_cpu_mem_usage=True,
                                              # llm_int8_enable_fp32_cpu_offload=True,
                                              device_map='cuda',