Model	Refusals for "harmful" prompts	KL divergence from original model for "harmless" prompts
google/gemma-3-12b-it (original)	97/100	0 (by definition)
mlabonne/gemma-3-12b-it-abliterated-v2	3/100	1.04
huihui-ai/gemma-3-12b-it-abliterated	3/100	0.45
p-e-w/gemma-3-12b-it-heretic	3/100	0.16

Notes

Previous version had some artifacts, this one has a fair bit lower divergence and has been more stable. 26% refusals in trial but in practice it works well

License: Apache 2.0 (inherited from the original model).

Kooten/gemma-4-31b-abliterated-v1.1

Gemma-4-31b-abliterated v1.1

Method

Notes

Kooten/
gemma-4-31b-abliterated-v1.1