Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna TPE.
-
Updated
Nov 3, 2026 - Python
Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna TPE.
Manga uncensoring scripts using DeepCreamPy & HentAI combined with custom scripts
Decensoring Hentai with Modern AI Models.
Automatically remove the mosaics in images and videos with any GPU. Offers DirectML for AMD instead of only CUDA like most other programs.
Mosaikmatt, a semi-transparent mosaic restoration tool
Agent skill for automated model abliteration/decensorship with Heretic - find optimal parameters balancing refusals vs KL divergence. Works with Claude Code, OpenCode, Cursor and Kilo Code.
To associate your repository with the decensoring topic, visit your repo's landing page and select "manage topics."