Token is a term used for any word or number or special character occurring in a sentence.
Source: Chapter 6, Section 6.3 / Text Normalisation – Tokenisation
The examiner expects the single correct term "Token." In tokenisation (a step of Text Normalisation in NLP), the text is broken into smaller units called tokens — these can be words, numbers, or special characters. Do not write "Punctuator," as that refers only to punctuation marks, not all types of units in a sentence.