Risk & governance
Interpretability
How directly a person can understand a model's internal mechanics.
Definition
Interpretability concerns whether model components and learned relationships can be inspected and related to understandable concepts.
Why it matters
Complex foundation models are often less inherently interpretable, increasing reliance on behavioural testing and external controls.
Related concepts
- Neural network
A layered mathematical model that learns relationships between inputs and outputs.
- Explainability
Methods that help people understand why a model produced an outcome.
- Model risk management
Governance for risks created by incorrect or misused models.