adding better error logs

app.py

@@ -6,11 +6,15 @@ import os
 
 ACCESS_TOKEN = os.getenv("HF_TOKEN")
 
+print("Access token loaded.")
+
 client = OpenAI(
     base_url="https://api-inference.huggingface.co/v1/",
     api_key=ACCESS_TOKEN,
 )
 
+print("OpenAI client initialized.")
+
 def respond(
     message,
     history: list[tuple[str, str]],
@@ -19,19 +23,27 @@ def respond(
     temperature,
     top_p,
 ):
+    print(f"Received message: {message}")
+    print(f"History: {history}")
+    print(f"System message: {system_message}")
+    print(f"Max tokens: {max_tokens}, Temperature: {temperature}, Top-P: {top_p}")
+
     messages = [{"role": "system", "content": system_message}]
 
     for val in history:
         if val[0]:
             messages.append({"role": "user", "content": val[0]})
+            print(f"Added user message to context: {val[0]}")
         if val[1]:
             messages.append({"role": "assistant", "content": val[1]})
+            print(f"Added assistant message to context: {val[1]}")
 
     messages.append({"role": "user", "content": message})
 
     response = ""
+    print("Sending request to OpenAI API.")
     
-    for message in  client.chat.completions.create(
+    for message in client.chat.completions.create(
         model="PowerInfer/SmallThinker-3B-Preview",
         max_tokens=max_tokens,
         stream=True,
@@ -40,12 +52,16 @@ def respond(
         messages=messages,
     ):
         token = message.choices[0].delta.content
-        
+        print(f"Received token: {token}")
         response += token
         yield response
+
+    print("Completed response generation.")
         
 chatbot = gr.Chatbot(height=600)
 
+print("Chatbot interface created.")
+
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -65,5 +81,8 @@ demo = gr.ChatInterface(
     chatbot=chatbot,
     theme="Nymbo/Nymbo_Theme",
 )
+print("Gradio interface initialized.")
+
 if __name__ == "__main__":
+    print("Launching the demo application.")
     demo.launch()