Heretic - Abliterated, Uncensored, Unrestricted POWER.
Collection
Models that have be abliterated using the HERETIC method. Done properly, this completely removed almost all censorship with no damage to the model. • 122 items • Updated • 63
Canz
Canzy
AI & ML interests
None yet
Recent Activity
upvoted a collection 9 days ago
Heretic - Abliterated, Uncensored, Unrestricted POWER. repliedto paulpham157's post 3 months ago
This isn't really related, but... do people generally prefer dark themes or light themes? repliedto paulpham157's post 3 months ago
This isn't really related, but... do people generally prefer dark themes or light themes?Organizations
upvoted a collection 9 days ago
repliedto paulpham157's post 3 months ago
Yessir, driving at night and those led headlights turning off all ability to see till the cars passed me. All I can do is try to stay straight and hope for the best. Lol
I'm exaggerating a little bit sometimes that's how it feels.
repliedto paulpham157's post 3 months ago
Dark, preferably black backgrounds with a blueish text that makes it look like its lit up. I'm not young anymore and had to start wearing glasses. I find this makes lettering standout without hurting my eyes.
liked a
model 4 months ago
reactedto giadap's post with 👍 6 months ago
Post
11094
One of the hardest challenges in AI safety is finding the right balance: how do we protect people from harm without undermining their agency? This tension is especially visible in conversational systems, where safeguards can sometimes feel more paternalistic than supportive.
In my latest piece for Hugging Face, I argue that open source and community-driven approaches offer a promising (though not exclusive) way forward.
✨ Transparency can make safety mechanisms into learning opportunities.
✨ Collaboration with diverse communities makes safeguards more relevant across contexts.
✨ Iteration in the open lets protections evolve rather than freeze into rigid, one-size-fits-all rules.
Of course, this isn’t a silver bullet. Top-down safety measures will still be necessary in some cases. But if we only rely on corporate control, we risk building systems that are safe at the expense of trust and autonomy.
Read the blog post here: https://huggingface.co/blog/giadap/preserving-agency
In my latest piece for Hugging Face, I argue that open source and community-driven approaches offer a promising (though not exclusive) way forward.
✨ Transparency can make safety mechanisms into learning opportunities.
✨ Collaboration with diverse communities makes safeguards more relevant across contexts.
✨ Iteration in the open lets protections evolve rather than freeze into rigid, one-size-fits-all rules.
Of course, this isn’t a silver bullet. Top-down safety measures will still be necessary in some cases. But if we only rely on corporate control, we risk building systems that are safe at the expense of trust and autonomy.
Read the blog post here: https://huggingface.co/blog/giadap/preserving-agency
- 7 replies
Well said. When I first started reading the initial post it kept going through my mind "your resonsible for protecting me? No thanks it's my responsibility to protect myself as it is everyone else." I come here to learn in all respects for my own personal interest and censoring what I am allowed to learn for "my best interest" just doesn't add up to me.
I was told when I was in my early 20s (mid 50s now) that if someone says something that pisses me off, I better go look in the mirror for why, and it's just true so I try to live by it.
reactedto jasoncorkill's post with 👍 about 1 year ago
Post
2869
Integrating human feedback is vital for evolving AI models. Boost quality, scalability, and cost-effectiveness with our crowdsourcing tool!
..Or run A/B tests and gather thousands of responses in minutes. Upload two images, ask a question, and watch the insights roll in!
Check it out here and let us know your feedback: https://app.rapidata.ai/compare
..Or run A/B tests and gather thousands of responses in minutes. Upload two images, ask a question, and watch the insights roll in!
Check it out here and let us know your feedback: https://app.rapidata.ai/compare