Claude 3 notices when a sentence about pizza toppings doesn't fit with its surrounding text. Whole internet including Tim Sweeney and Margaret Mitchell concludes that it's probably self-aware now.

mozz@mbin.grits.dev · 9 months ago

Claude 3 notices when a sentence about pizza toppings doesn't fit with its surrounding text. Whole internet including Tim Sweeney and Margaret Mitchell concludes that it's probably self-aware now.

kinttach · 9 months ago

The Anthropic researcher didn’t have this take. They were just commenting that it was interesting. It’s everyone else who seemed to think it meant something more.

Doesn’t it just indicate that the concept of needle-in-a-haystack testing is included in the training set?

Claude 3 notices when a sentence about pizza toppings doesn't fit with its surrounding text. Whole internet including Tim Sweeney and Margaret Mitchell concludes that it's probably self-aware now.

Claude 3 notices when a sentence about pizza toppings doesn't fit with its surrounding text. Whole internet including Tim Sweeney and Margaret Mitchell concludes that it's probably self-aware now.

Anthropic’s Claude 3 causes stir by seeming to realize when it was being tested