Modelling hierarchical intent structure

I have an intent chitchat in my domain, and I want to classify it into more specific intent such as questions about name, questions about age. Is there any way for me to perform this. I want these intent to be a sub-intent of chitchat since I don’t want the chatbot to accidentally identify these intents in the formal context

yes! check out this tutorial Integrate response retrieval models in assistants built with Rasa

cool, I will try it out

Can you tell me how to move to a new line when writing response

That completely depends on the front-end of your bot. if you’re using custom front-end on a web app then a <br> tag could be added in the response.
In case of a command line based bot response you can add \n in you text.
Or you can edit your domain file for templates with newlines in them like:

utter_some_message:
    - text: >
        One
        Two
        Three

I’ve tried both ways for command line based bot response, and it doesn’t work

In case of command line a \n break should work.
For example in the default example bot provided by running rasa init, if you change the domain file template to

templates:
  utter_greet:
  - text: "Hey! \nHow are you?"

And then go ahead and retrain your model, the output you’ll get is:

Your input ->  hi                                                               
Hey! 
How are you?
Your input -> 

But I see the retrieval response is not the same as utterrance, and I write the responses in responses.md with the same format mentioned in the blog post

responses.md (3.4 KB) I tried every means