On: Anthropic publishes the 'system prompts' that make...

[MacsHeadroom]

Anthropic/Claude does not use any RLHF.

[teqsun → MacsHeadroom]

Is that a claim they've made or has that been externally proven?

[cjbillington → MacsHeadroom]

What do they do instead? Given we're not talking to a base model.

[tqi → cjbillington]

Supposedly they use "RLAIF", but honestly given that the first step is to "generate responses... using a helpful-only AI assistant" it kinda sounds like RLHF with more steps.

https://www.anthropic.com/research/constitutional-ai-harmles...