Update app.py

app.py

@@ -71,8 +71,8 @@ def generator(input_ids, generation_config, max_new_tokens):
 def loadModel():
     global model, tokenizer
     if model is None:
-        #from llama_rope_scaled_monkey_patch import replace_llama_rope_with_scaled_rope
-        #replace_llama_rope_with_scaled_rope()
+        from llama_rope_scaled_monkey_patch import replace_llama_rope_with_scaled_rope
+        replace_llama_rope_with_scaled_rope()
         model = transformers.AutoModelForCausalLM.from_pretrained(
                     base_model,
                     torch_dtype=torch.float16,