Update app.py

app.py

@@ -6,9 +6,18 @@ import os
 # Load model and tokenizer with the token from environment variables
 model_name = "meta-llama/Llama-2-7b-hf"
 token = os.getenv("HUGGINGFACE_TOKEN")  # Get token from environment
+
+# Add print statements for debugging
+print("Loading tokenizer...")
 tokenizer = AutoTokenizer.from_pretrained(model_name, token=token)
+print("Tokenizer loaded.")
+
+print("Loading model...")
 model = AutoModelForCausalLM.from_pretrained(model_name, token=token, torch_dtype=torch.float16)
+print("Model loaded.")
+
 model = model.to("cuda" if torch.cuda.is_available() else "cpu")
+print("Model moved to device.")
 
 # Function to generate responses
 def generate_response(user_input, chat_history):
@@ -29,3 +38,7 @@ def chat_interface():
         response, chat_history = generate_response(user_input, chat_history)
         return response
     gr.Interface(fn=respond, inputs="text", outputs="text", title="LLaMA-2 Chatbot").launch()
+
+# Call the interface function to start the app
+print("Launching Gradio interface...")
+chat_interface()