debug

app.py

@@ -8,7 +8,6 @@ if torch.cuda.is_available():
     device = "cuda"
 else:
     device = "cpu"
-
 device_map={'': 0}
 def generate_instruction_prompt(instruction, input=None):
     if input:
@@ -71,6 +70,7 @@ def load_lora(lora_path, base_model="decapoda-research/llama-7b-hf"):
         device_map=device_map,
         low_cpu_mem_usage=True,
     )
+    print("Loading LoRA...")
     lora = PeftModel.from_pretrained(
         model,
         lora_path,