update with gpt2

app.py

@@ -61,7 +61,7 @@ def respond(message, history, system_message, max_tokens, temperature, top_p):
 
         for chunk in client.text_generation(
             prompt,
-            model="microsoft/Phi-3-mini-4k-instruct",   # ✅ causal LM
+            model="gpt2",   # ✅ causal LM
             max_new_tokens=max_tokens,
             temperature=temperature,
             top_p=top_p,