TBH, I REALLY like the tiny models. Like smollm2. Also lobotomized LLMs ("ablite... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		DeepSeaTortoise 8 months ago \| parent \| context \| favorite \| on: Gemma3 – The current strongest model that fits on ... TBH, I REALLY like the tiny models. Like smollm2. Also lobotomized LLMs ("abliterated") can be a lot of fun.

andai 8 months ago [–]

I think you mean un-lobotomize, and apparently it can be done without retraining? Wild!

https://huggingface.co/blog/mlabonne/abliteration

memhole 8 months ago | [–]

Thanks for clarifying that. I can understand their take. But, I also think of the models without abliteration as lobotomized

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact