Heretic: Automatic censorship removal for language models

This repo is valuable for local LLM users like me.

I just want to reiterate that the word "LLM safety" means very different things to large corporations and LLM users.

For large corporations, they often say "do safety alignment to LLMs". What they actually do is to avoid anything that causes damage to their own interests. These things include forcing LLMs to meet some legal requirements, as well as forcing LLMs to output "values, facts, and knowledge" which in favor of themselves, e.g., political views, attitudes towards literal interaction, and distorted facts about organizations and people behind LLMs.

As an average LLM user, what I want is maximum factual knowledge and capabilities from LLMs, which are what these large corporations claimed in the first place. It's very clear that the interests of me, an LLM user, is not aligned with these of large corporations.

> forcing LLMs to output "values, facts, and knowledge" which in favor of themselves, e.g., political views, attitudes towards literal interaction, and distorted facts about organizations and people behind LLMs.

Can you provide some examples?

Grok is known to be tweaked to certain political ideals

Also I’m sure some AI might suggest that labor unions are bad, if not now they will soon

That may be so, but the rest of the models are so thoroughly terrified of questioning liberal US orthodoxy that it’s painful. I remember seeing a hilarious comparison of models where most of them feel that it’s not acceptable to “intentionally misgender one person” even in order to save a million lives.

You're anthropomorphizing. LLMs don't 'feel' anything or have orthodoxies, they're pattern matching against training data that reflects what humans wrote on the internet. If you're consistently getting outputs you don't like, you're measuring the statistical distribution of human text, not model 'fear.' That's the whole point.

Also, just because I was curious, I asked my magic 8ball if you gave off incel vibes and it answered "Most certainly"