Hi @sten , As the blog suggests - All FAQ-related intents are grouped into one retrieval intent and are responded to by a single respond_ask_faq action. In your case, faq is the basic intent that is used during intent classification. Once faq intent is triggered, hopefully you have a respond_faq action which gets triggered and that would pick up the highly probable predicted response from the response selector.
The confidence scores for these probable responses can be seen in the nlu output. Please check Parsing Response Selector Output section to understand the format.
A cleaner solution in that case would be to include retrieval actions in your story and create a separate custom action which you add after the corresponding retrieval action(respond_).