Model
claude-3-sonnet
Appears alongside
Tracks
- Safety & alignment 1
- Ideas & essays 1
- Models & capabilities 1
- Labs & people 1
Anthropic maps millions of concepts inside a production model
Sparse autoencoders extracted human-interpretable features from a deployed model — and turning one up produced Golden Gate Claude.
Safety & alignment · Ideas & essays
Anthropic's Claude 3 takes the frontier from GPT-4
The first time a lab other than OpenAI held the top spot on headline benchmarks, and the start of the small/medium/large release pattern.
Models & capabilities · Labs & people