AI Model documentation, also known as a “Model Card,” provides essential information about a model’s characteristics, performance, and limitations. The purpose of a Model Card is to ensure the AI model is transparent, accessible, and understandable. A well-crafted Model Card is written in clear language and typically covers the following key areas:
- Model Overview
- Training Data
- Evaluation Metrics
- Intended Use and Limitations
- Model Fairness and Bias
- Safety and Reliability
- Version History and Updates
- Deployment Details
As Google explains:
“A model card for a language translator, for example, may provide guidance around jargon, slang, and dialects, or measure its tolerance for differences in spelling. There are many forms that transparent documentation can take, and we encourage a flexible approach that allows for variation in model type and evaluation specifics.”
Model Cards are a critical tool for supporting Responsible AI by fostering greater transparency, accountability, fairness, trust, security, and governance in AI development and deployment.
Source
https://modelcards.withgoogle.com/about
Eryk Salvaggio. “A Critical Intro to NLP & LLMs.” Masters in Design for Responsible Artificial Intelligence, Nov 5 2024, Elisave, Online. reading https://github.com/meta-llama/llama/blob/main/MODEL_CARD.md