update

app.py

@@ -27,19 +27,19 @@ def respond(
     max_tokens,
     temperature,
 ):
-    prompt = """<|start_header_id|>system<|end_header_id|>
+    # prompt = """<|start_header_id|>system<|end_header_id|>
 
-    You are a helpful assistant.<|eot_id|><|start_header_id|>user<|end_header_id|>
-    {message}
-    <|eot_id|><|start_header_id|>assistant<|end_header_id|>
-    """
-    model_inputs = generate_custom_mask(tokenizer, [prompt], device)
+    # You are a helpful assistant.<|eot_id|><|start_header_id|>user<|end_header_id|>
+    # {message}
+    # <|eot_id|><|start_header_id|>assistant<|end_header_id|>
+    # """
+    # model_inputs = generate_custom_mask(tokenizer, [prompt], device)
    
-    outputs = model.generate(temperature=0.7, max_tokens=64, **model_inputs)
-    outputs = outputs[:, model_inputs['input_ids'].shape[1]:]
-    result = tokenizer.batch_decode(outputs, skip_special_tokens=True)
+    # outputs = model.generate(temperature=0.7, max_tokens=64, **model_inputs)
+    # outputs = outputs[:, model_inputs['input_ids'].shape[1]:]
+    # result = tokenizer.batch_decode(outputs, skip_special_tokens=True)
 
-    return result, []
+    return "test", []
 
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface