SESKStuff Everyone Should Know
SESK: Stuff Everyone Should Know · sesk.info/ai-parameters-versus-tokens

Which statement correctly distinguishes model parameters from tokens?

Parameters are learned values, while tokens are processed units.

Start playing free

  1. Both are temporary prompts
  2. Tokens are learned weights
  3. Parameters are learned, tokens are inputs or outputs ✔

Why: Weights and biases are parameters retained in a trained model. Tokens are the atomic units the language model reads, predicts, or generates.

Huh, didn't know that: A model can process many different token sequences using the same fixed set of parameters.

The dad joke
Parameters keep their values, while tokens come and go.
Like this card?
Play the game

References

Answer: Parameters are learned, tokens are inputs or outputs. Weights and biases are parameters retained in a trained model. Tokens are the atomic units the language model reads, predicts, or generates.

  1. Google Machine Learning Glossary developers.google.com/machine-learning/glossary
    parameters are the various weights and bias that the model learns during training
    Checked 2026-10-10.
  2. Google Machine Learning Crash Course: Introduction to LLMs developers.google.com/machine-learning/crash-course/llm
    Tokens are the atomic unit or smallest unit of language modeling.
    Checked 2026-10-10.

Think this answer is wrong? Tap "Challenge this answer" on the card in the game and tell us why.

Test yourself in the game

Artificial IntelligenceAbout how many parameters did AlexNet, the 2012 ImageNet breakthrough network, have?Artificial IntelligenceWhich of these is a hyperparameter you set, not something the model learns?Artificial IntelligenceWhat can a neuron's bias change even when its inputs stay fixed?

That's stuff everyone should know.