Update app.py

app.py

@@ -70,7 +70,7 @@ def generator(input_ids, generation_config, max_new_tokens):
 def loadModel():
     from llama_rope_scaled_monkey_patch import replace_llama_rope_with_scaled_rope
     replace_llama_rope_with_scaled_rope()
-    base_model = transformers.AutoModelForCausalLM.from_pretrained(
+    t_model = transformers.AutoModelForCausalLM.from_pretrained(
                 base_model,
                 torch_dtype=torch.float16,
                 cache_dir=cache_dir,
@@ -78,7 +78,7 @@ def loadModel():
             )
     
     model = PeftModel.from_pretrained(
-                base_model,
+                t_model,
                 lora_weights,
                 device_map="auto",
                 cache_dir=cache_dir,