AI Fundamentals: How Large Language Models Actually Work
by Froga
Tokens, parameters, attention, hallucination — the mechanics behind ChatGPT and every model like it, explained once so the vocabulary stops being intimidating.
You're previewing this room. Log in or create a free account to answer the questions and track your progress.
Study material
Tokens can be whole words or just fragments of one — "unbelievable" might become three tokens. The model always works over a sequence of tokens; it never reads raw characters the way a person does.
Before generating a reply, a large language model first breaks the input text into small pieces called...
multiple choice
tokens
packets
vectors
bytes
Ready to test yourself?
Sign up free to answer, score points, and build your streak.