huihui-ai

/

Huihui-gpt-oss-120b-BF16-abliterated

like 35

I was wondering if you would be willing to share the method / release the code you use to ablate models. This one works fairly well and from my testing so far, maintains coherence in parity with the original. I'm still getting some refusals here and there though and I would like to use your code as a starting point to ablate it further.