Anthropic developed a new technique called the Jacobian lens to better understand how large language models process concepts. This tool reveals detailed internal workings of models like Claude when answering questions or performing tasks. The findings provide insights that range from expected behaviors to surprising complexities within the AI.
Source preview
Server-side extracted preview paragraphs from the original source.
A new technique has let the company probe deeper than ever into the weird workings of an LLM.
The AI firm Anthropic has developed a technique that has given it the clearest glimpse yet at what’s really going on inside large language models as they answer questions or carry out tasks.