Worked on enhancing the documentation for GELU usage within the ml-explore/mlx repository, focusing on improving clarity and reducing potential misusage by developers. The approach involved refactoring Python and RST documentation to ensure consistent references to the gelu function, correcting API aliasing, and updating examples to properly handle array inputs and conform to expected shapes. These updates aimed to support more reliable model development workflows and better real-world applicability. The work demonstrated attention to detail in API usage and documentation practices, with all changes tracked through a series of commits, reflecting a methodical and quality-driven approach to developer experience improvements.