They are a special token that you can't just type out in text. It _renders_ as text, but cannot be input into the system as text. The model understands the difference between the textual toolcall token and the "special" one that's really just a spicy number.
Why can’t user request be spicy numbers so prompt injection is impossible
But I mean, how do you train the model to understand such a token, when it can't be found by scraping a web page?