I got a straight answer out of QwQ-LCoT-7B-Instruct.Q4_K_M five out of five times for the Tiananmen Square question.

I just want to mess with this one too. I had a hard time finding an abliterated one before that didn’t fail the Tiananmen Square question regularly.

Can’t wait to try a distillation. The full model is huge.

Computers are female in Spanish and in English too, I suspect. It used to be a job mostly held by women.

Hybrid animal illustrations, abstract art, portraits, and world morph images where you illustrate something made out of an unusual material.

But there’s already tons of generative art that gets discussed, lauded, shared, and remixed. Even when in negative connotations, it’s proving that generative art, warts and all, is a new form of art that is shaking things up, challenging preconceptions, and getting people angry - just like art should.

Yeah, it’s distilled from deepseek and abliterated. The non-abliterated ones give you the same responses as Deepseek R1.

I’ve heard of this happening when you generate datasets with ChatGPT to help train your model. OpenAI doesn’t want you doing this, making it against their terms of use, but there’s nothing they can do to stop people. You can generate some really good synthetic datasets from ChatGPT, and it’s perfectly legal to do.

Were you running it locally?

The answer I got out of DeepSeek-R1-Distill-Llama-8B-abliterate.i1-Q4_K_S

deleted by creator

I think this is the other one I remember seeing.

Have you seen this? There was another paper, but I can’t remember the name of it right now.

Matt wasn’t that good at tennis.