Update app.py

app.py

@@ -3,15 +3,16 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 import gradio as gr
 
 # Adjust this to your model ID
-model_id = "decision-oaif/Meta-Llama-3-8B-Instruct-sft-alfworld-iter1"
-#peft_model_id = "decision-oaif/Meta-Llama-3-8B-Instruct-sft-intercode-python-iter0"
+model_id = "meta-llama/Meta-Llama-3-8B-Instruct"
+
+peft_model_id = "decision-oaif/Meta-Llama-3-8B-Instruct-sft-intercode-bash-iter0"
 # Load model with device map and dtype
 model = AutoModelForCausalLM.from_pretrained(
     model_id,
     torch_dtype=torch.bfloat16,
     device_map="auto"
 )
-#model.load_adapter(peft_model_id)
+model.load_adapter(peft_model_id)
 
 # Load tokenizer and set truncation and padding
 tokenizer = AutoTokenizer.from_pretrained(model_id, truncation=True, padding=True)
@@ -60,4 +61,4 @@ def generate_response(messages):
 iface = gr.Interface(fn=generate_response, inputs="json", outputs="text", title="Meta-Llama-3-8B-Instruct")
 
 # Launch the interface
-iface.launch()
+iface.launch()