I assume it's done with fine tuning. Meaning you come up with responses that capture topics you want to avoid, then let the model convert these unwords into th

I assume it's done with fine tuning. Meaning you come up with responses that capture topics you want to avoid, then let the model convert these unwords into their corresponding latent space, and train it to exclude that space altogether.