Suitable for response testing

app.py

@@ -81,7 +81,7 @@ def format_aura_response(raw_response):
     
     return raw_response
 
-def respond(message, history):
+def respond(message, history, max_length=150, temperature=0.9, top_p=0.9, top_k=50, repetition_penalty=1.2):
     """Generate response for the chatbot with Aura personality"""
     try:
         # Crisis detection - highest priority
@@ -103,16 +103,16 @@ def respond(message, history):
         # Tokenize
         input_ids = tokenizer.encode(conversation, return_tensors="pt")
         
-        # Generate response with improved settings for more creative, less repetitive responses
+        # Generate response with configurable parameters
         with torch.no_grad():
             chat_history_ids = model.generate(
                 input_ids,
-                max_length=150,
-                temperature=0.9,
-                top_p=0.9,
-                repetition_penalty=1.2,
+                max_length=max_length,
+                temperature=temperature,
+                top_p=top_p,
+                repetition_penalty=repetition_penalty,
                 do_sample=True,
-                top_k=50,
+                top_k=top_k,
                 pad_token_id=tokenizer.eos_token_id,
                 no_repeat_ngram_size=3
             )
@@ -290,18 +290,49 @@ with gr.Blocks(title="Aura - Your Supportive Friend") as demo:
     with gr.Row():
         clear = gr.Button("Clear Chat", variant="secondary")
     
+    # Add parameter controls with Aura-friendly labels
+    with gr.Accordion("⚙️ Response Settings (Advanced)", open=False):
+        gr.Markdown("*Adjust these settings to change how Aura responds. Default values work well for most conversations.*")
+        with gr.Row():
+            max_length = gr.Slider(
+                minimum=50, maximum=1500, value=150, step=10,
+                label="Response Length",
+                info="How detailed Aura's responses are"
+            )
+            temperature = gr.Slider(
+                minimum=0.1, maximum=2.0, value=0.9, step=0.1,
+                label="Creativity",
+                info="Higher values make responses more creative"
+            )
+        with gr.Row():
+            top_p = gr.Slider(
+                minimum=0.1, maximum=1.0, value=0.9, step=0.05,
+                label="Focus",
+                info="How focused responses are (lower = more focused)"
+            )
+            top_k = gr.Slider(
+                minimum=10, maximum=100, value=50, step=5,
+                label="Word Choice Variety",
+                info="Range of words Aura considers"
+            )
+        repetition_penalty = gr.Slider(
+            minimum=1.0, maximum=2.0, value=1.2, step=0.1,
+            label="Repetition Control",
+            info="Prevents Aura from repeating phrases"
+        )
+    
     def user(user_message, history):
         return "", history + [[user_message, None]]
     
-    def bot(history):
+    def bot(history, max_len, temp, top_p_val, top_k_val, rep_penalty):
         if history and history[-1][1] is None:
             user_message = history[-1][0]
-            bot_response = respond(user_message, history[:-1])
+            bot_response = respond(user_message, history[:-1], max_len, temp, top_p_val, top_k_val, rep_penalty)
             history[-1][1] = bot_response
         return history
     
     msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
-        bot, chatbot, chatbot
+        bot, [chatbot, max_length, temperature, top_p, top_k, repetition_penalty], chatbot
     )
     clear.click(lambda: None, None, chatbot, queue=False)