← Back to context

Comment by bashbjorn

17 hours ago

You're right, there must be a good and simple way to do it.

Obviously the prefix-with-backslash convention won't do it. The escaping system could be something like inserting a character on the second position in the text repr, and reversing that on output too if it matches an escaped known special token.

Changing the vocab on the fly requires tokenizing things separately, breaking the chat template.

Anecdotally, even claude code has an anneurism sometimes when listing special tokens. Idk exactly what claude's <eos> token is, but I'm fairly sure I've seen it stop generation when it tried to generate it before.

I should also say that I've (clearly) not thought about this deeply. There should be a simpler way to do it.