No One Truly Knows How AI Systems Work. A New Discovery Could Change That
Researchers developed a technique for essentially scanning the “brain” of an AI model, allowing them to identify collections of neurons—called “features”—corresponding to different concepts. And for the first time, they successfully used this technique on a frontier large language model, Anthropic’s Claude Sonnet, the lab’s second-most powerful system.
EIN Presswire does not exercise editorial control over third-party content provided, uploaded, published, or distributed by users of EIN Presswire. We are a distributor, not a publisher, of 3rd party content. Such content may contain the views, opinions, statements, offers, and other material of the respective users, suppliers, participants, or authors.
