Two-stage classifier
Named by 2 kept items in this workspace.
Search every workspace for Two-stage classifier. It matches the name as written, so it can find other things that share it.
The items that name it 2
Safeguard of the kind described in constitutional classifiers research that detects and blocks attempts to extract Claude's hidden reasoning; false positive rate under 0.1%
Anthropic safeguard system, of the kind described in its constitutional classifiers research, that detects and blocks attempts to extract Claude's hidden reasoning. False positive rate is under 0.1%.
All technologies in this workspace.