a team led by data and computer scientists at the University of California San Diego has given neural networks the equivalent of an X-ray to uncover how they actually learn. The researchers found that a formula used in statistical analysis provides a streamlined mathematical description of how neural networks, such as GPT-2, learn relevant patterns in data, known as features. This formula also explains how neural network use these relevant patterns to make predictions. One of the reasons neural networks are considered a black box is because it is often difficult to describe