Update app.py

app.py

@@ -192,10 +192,15 @@ with gr.Blocks(css=MODEL_SELECTION_CSS, theme='gradio/soft') as demo:
             "last conversation, • clean the chat history, • dynamic kick-starting prompt templates, • adjusting text generation parameters, "
             "• inspecting the actual prompt that the model sees. The underlying Large Language Model is the [Meta AI](https://ai.meta.com/)'s "
             "[LLaMA2-70B](https://huggingface.co/meta-llama/Llama-2-70b-chat-hf) which is hosted as [Hugging Face Inference API](https://huggingface.co/inference-api), "
-            "and [Text Generation Inference](https://github.com/huggingface/text-generation-inference) is the underlying serving framework. "
-            "\n***NOTE:*** If you are subscribing [PRO](https://huggingface.co/pricing#pro), you can simply duplicate this space and use your "
-            "Hugging Face Access Token to run the same application. Just add `HF_TOKEN` secret with the Token value accorindg to [this guide](https://huggingface.co/docs/hub/spaces-overview#managing-secrets-and-environment-variables)",
+            "and [Text Generation Inference](https://github.com/huggingface/text-generation-inference) is the underlying serving framework. ",
             elem_classes=["center"])
+        )
+        gr.Markdown(
+            "***NOTE:*** If you are subscribing [PRO](https://huggingface.co/pricing#pro), you can simply duplicate this space and use your "
+            "Hugging Face Access Token to run the same application. Just add `HF_TOKEN` secret with the Token value accorindg to [this guide](https://huggingface.co/docs/hub/spaces-overview#managing-secrets-and-environment-variables)",
+            elem_classes=["center"]
+        )
+            
         
         with gr.Row():
             with gr.Column(scale=1, min_width=180):