Update app.py

app.py

@@ -18,27 +18,27 @@ def respond(
     temperature,
     top_p,
 ):
-# 2. Construct the Prompt (Crucial!)
-prompt = f"{system_message}\n" 
-for user_msg, bot_msg in history:
-    prompt += f"User: {user_msg}\nAssistant: {bot_msg}\n"
-prompt += f"User: {message}\nAssistant:"
-
-# 3. Generate with the Pipeline
-try:
-    response = pipe(
-        prompt,
-        max_new_tokens=max_tokens,
-        temperature=temperature,
-        top_p=top_p,
-    )[0]["generated_text"]
-    #Extract the bot's reply (adjust if your model format is different)
-    bot_response = response.split("Assistant:")[-1].strip()
-    yield bot_response
-
-except Exception as e:
-    print(f"Error during generation: {e}")
-    yield "An error occurred during generation." #Handle generation errors.
+    # 2. Construct the Prompt (Crucial!)
+    prompt = f"{system_message}\n" 
+    for user_msg, bot_msg in history:
+        prompt += f"User: {user_msg}\nAssistant: {bot_msg}\n"
+    prompt += f"User: {message}\nAssistant:"
+    
+    # 3. Generate with the Pipeline
+    try:
+        response = pipe(
+            prompt,
+            max_new_tokens=max_tokens,
+            temperature=temperature,
+            top_p=top_p,
+        )[0]["generated_text"]
+        #Extract the bot's reply (adjust if your model format is different)
+        bot_response = response.split("Assistant:")[-1].strip()
+        yield bot_response
+    
+    except Exception as e:
+        print(f"Error during generation: {e}")
+        yield "An error occurred during generation." #Handle generation errors.
 
 # 4. Gradio Interface (No changes needed here)
 demo = gr.ChatInterface(