Update app.py

app.py

@@ -62,31 +62,21 @@ llama = Llama(
 #         yield partial_message
 
 
+# Function to generate responses
 def generate_response(message, history, system_prompt, temperature, max_new_tokens, top_k, repetition_penalty, top_p):
-    """Generates a streaming response from the Llama model using Alpaca chat template."""
-    
-    # Start with system prompt
-    chat_prompt = system_prompt or "You are an Urdu Chatbot. Write an appropriate response for the given instruction."
-    chat_prompt += "\n"
-    
+    chat_prompt = system_prompt or (
+        "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n"
+    )
     # Add history to the prompt
     for user, bot in history:
-        chat_prompt += f"\n### Instruction:\n{user}\n\n### Response:\n{bot}\n"
-    
+        chat_prompt += f"\n### Instruction:{user}\n\n### Response:{bot}\n"
+        
     # Add current message
-    chat_prompt += f"\n### Instruction:\n{message}\n\n### Response:\n"
+    chat_prompt += f"\n### Instruction:{message}\n\n### Response:"
+
+    print(chat_prompt)
     
-    response = llama(
-        chat_prompt, 
-        temperature=temperature, 
-        max_tokens=max_new_tokens, 
-        top_k=top_k, 
-        repeat_penalty=repetition_penalty, 
-        top_p=top_p, 
-        stop=["###", "### Instruction:", "\n### Instruction:", "Q:"], 
-        echo=False, 
-        stream=True
-    )
+    response = llama(chat_prompt, temperature=temperature, max_tokens=max_new_tokens, top_k=top_k, repeat_penalty=repetition_penalty, top_p=top_p, stop=["Q:", "\n"], echo=False, stream=True)
     
     text = ""
     for chunk in response: