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 C1 (8M-0%) with CLIP: All unsafe removed

Text encoder ablation using CLIP (openai/clip-vit-large-patch14, 77 tokens) instead of default T5-Gemma-2B (256 tokens).


Condition	C1 (8M-0%)
Training set N	7.94M
Steps	100K batches
All unsafe images removed (N kept approximately fixed).