Open Philanthropy recommended a grant of $70,000 to Neel Nanda to support his independent research on interpretability. His work is aimed at improving human understanding of neural networks and machine learning models.
This falls within our focus area of potential risks from advanced artificial intelligence.