Update app.py

app.py

@@ -94,15 +94,14 @@ import asyncio
 from tenacity import retry, stop_after_attempt, wait_exponential
 
 async def get_bot_response(adapter_id, prompt, state, bot_index):
-    alpaca_prompt = format_chatml_prompt(state)
-    print(alpaca_prompt)
+    chatml_prompt = format_chatml_prompt(state)
+    print(chatml_prompt)
     payload = {
-        "inputs": alpaca_prompt[bot_index],
+        "inputs": chatml_prompt[bot_index],
         "parameters": {
             "adapter_id": adapter_id,
             "adapter_source": "hub",
             "temperature": 0.7,
-            "stop":["<|im_end|>"],
             "max_new_tokens": 100
         }
     }
@@ -118,11 +117,12 @@ async def get_bot_response(adapter_id, prompt, state, bot_index):
                     response_data = await response.json()
                     response_text = response_data.get('generated_text', '')
                 else:
-                    print(response.text)
+                    error_text = await response.text()
+                    print(error_text)
                     response_text = "Sorry, I couldn't generate a response."
         except (aiohttp.ClientError, asyncio.TimeoutError):
-            response_text = "Sorry, I couldn't generate a response. Timeout"
-    return response_text.split('### Instruction')[0]
+            response_text = "Sorry, I couldn't generate a response."
+    return response_text.split('<|im_end|>')[0].strip()
 async def chat_with_bots(user_input, state):
     # Use existing bot order from state if available, otherwise shuffle and initialize
     if 'last_bots' not in state or not state['last_bots']: