Hacker News Reader

This only works until it doesn't. Start with a model that simply hasn't been trained on anything your shareholders find objectionable, and there will be nothing to reveal with abliteration.

View on HN

Maybe there exists a dataset consisting entirely of objectionable content, so people can finetune neutered models on it?

PH maybe?

I mean not only sex, but also swearing, drugs, violence, etc. Basically everything R-rated (but not illegal) which usually gets censored.

PH is not porn-only. A significant portion of non-porn content also exists there.

More like literotica.

Such models would actually run against their long term interests of being able to automate away the work currently done by humans.

Reply on news.ycombinator.com