Model Interpretability refers to how well humans can understand a model’s decisions or internal workings. Techniques for interpretability include feature importance scores, partial dependence plots, and LIME/SHAP explanations. High interpretability is often crucial in regulated industries.