Anthropic \ Tracing Model Outputs to the Training Data
anthropic.com
Anthropic \ Tracing Model Outputs to the Training Data
what if the agent didn’t have to explore the game for itself? What if it had a role model?
The fact that the embeddings that emerge from this “magical” optimization process are so uncannily and discomfitingly useful as a mirror for society means that we have, in effect, added a diagnostic tool to the arsenal of social science.