What is a token in a foundation model?
Why A is the answer
A small unit of text that the model processes
Text is split into tokens, and usage is commonly priced by them.
In detail
- A token is a small unit of text.
- A token is often a word or part of a word.
Use case
An illustrative scenario, not a real incident.
A sentence is split into tokens before a model reads it. A team estimates its monthly bill from token counts.
See it in different ways
flowchart LR A["Data selection"] --> B["Model selection"] B --> C["Pre-training"] C --> D["Fine-tuning"] D --> E["Evaluation"] E --> F["Deployment"] F --> G["Feedback"]
mindmap
root((Generative AI))
Token
Unit of text
Drives cost
Embedding
Meaning as numbers
Chunking
Focused passages
Hallucination
Confident and false
Temperature
Varietyflowchart LR S0["Sentence"] --> S1["Split into tokens"] --> S2["Model processes them"]
Key point: Tokens are units of text.
Why the other options are wrong
- B. A security credential for the AWS account
- Here it does not mean a credential.
- C. A full document in the training data
- A document holds many tokens.
- D. A payment method used for Amazon Bedrock charges on each invoice
- It is not a payment method.