Transformers and LLMs

Token, explained

A token is a unit of text, code, audio, or other input representation that a model processes.
Back to Glossary

Part of the Semantic Notion technical glossary.

Short Definition

A token is a unit of text, code, audio, or other input representation that a model processes.

Intuition

Models do not usually read raw words exactly as humans do. They read chunks produced by a tokenizer or modality encoder.

Technical Definition

A token is a discrete element from a vocabulary or encoded stream that is mapped to an embedding before model processing.

Example

The word "unbelievable" might be split into multiple subword tokens.

Common Misunderstandings

A token is not always a full word.
Token counts affect cost, latency, and context length.

Same Category

More from Transformers and LLMs

AttentionAttention is a neural network mechanism that lets a model decide which parts of an input are most relevant when producing a representation or output.Context WindowA context window is the maximum amount of input and generated tokens a model can consider at one time.Cross-AttentionCross-attention lets one sequence attend to information from a different sequence or modality.Hybrid AttentionHybrid attention combines different attention patterns, such as local and global attention, inside one model.KV CacheA KV cache stores attention keys and values from previous tokens so autoregressive generation does not recompute them every step.Multimodal ModelA multimodal model can process or generate more than one kind of data, such as text, images, audio, or video.

Related Reading

Essays that use this concept

Transformers and LLMsWhat Is Attention in Transformers? Explained Intuitively

Start here

Need the broader concept map?

Return to the glossary index to move from this definition into adjacent AI, ML, mathematics, and computer science terms.

Browse Glossary