view post Post 7751 āļø AutoAbliterationI made a Colab notebook to automatically abliterate models.It's quite general, so you can do interesting stuff like blocking a given language in the model outputs.š» Colab: https://colab.research.google.com/drive/1RmLv-pCMBBsQGXQIM8yF-OdCNyoylUR1?usp=sharing See translation š 30 30 š„ 15 15 š 5 5 ā¤ļø 4 4 š 2 2 + Reply
view post Post 6067 āļø Gemma 3 AbliteratedI noticed that Gemma 3 was much more resilient to refusal removal than other models like Qwen 2.5.I experimented with different recipes and improved the abliteration technique I wrote about last year.It's still experimental but the refusal rate is super low in my tests. Enjoy! mlabonne/gemma-3-4b-it-abliterated mlabonne/gemma-3-12b-it-abliterated mlabonne/gemma-3-27b-it-abliterated See translation 1 reply Ā· š 14 14 š 9 9 š 3 3 ā¤ļø 2 2 + Reply
huihui-ai/DeepSeek-R1-Distill-Qwen-Coder-32B-Fusion-9010 Text Generation ā¢ Updated Feb 22 ā¢ 87 ā¢ 7
huihui-ai/Llama-3.3-70B-Instruct-abliterated Text Generation ā¢ Updated Dec 12, 2024 ā¢ 62.6k ā¢ 31
netcat420/MFANN-Llama3.1-Abliterated-SLERP-TIES-V3 Text Generation ā¢ Updated Nov 26, 2024 ā¢ 5 ā¢ 2