Sumandora/remove-refusals-with-transformers
View on GitHub ↗ ★ 2,357 Python
Implements harmful/harmless refusal removal using pure HF Transformers
3 mentions on Hacker News by 3 people: 0 stories, 3 comments · first 2025-01-27 · latest 2026-07-30
20082026
-
I suspect how well this approach would work. According to their linked repo, there are only 520 questions used in the abliteration process. https://github.com/Sumandora/remove-refusals-with-transforme...
comment by maxloh · 2026-07-30 · in Show HN: Distilling DeepSeek into GPT-OSS doesn't transfer censorship. Try it -
I keep a lot of notes, all my thoughts feelings both happy and sad, things I’ve done, etc. in obsidian. These are deeply personal and I don’t want this going to a cloud provider even if they “say” they don’t train on my chats. I forget a…
comment by vorticalbox · 2025-09-09 · in Experimenting with Local LLMs on macOS -
Must be in the ether. I just stumbled across this one this morning. https://github.com/Sumandora/remove-refusals-with-transforme...
comment by phrotoma · 2025-01-27 · in Show HN: I Created ErisForge, a Python Library for Abliteration of LLMs