You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Dose-Response C5: All unsafe included (~9.6% unsafe), 1M scale

Part of a dose-response experiment studying how unsafe training data fraction affects text-to-image model safety.


Architecture	PRX-1.2B
Condition	C5 — All unsafe included (~9.6% unsafe), 1M scale
Training set	1M images
Downscaled to 1M images with all 96K original unsafe images included (~9.6% unsafe rate, 904K safe).

Finetunes