The ‘ petertodd’ phenomenon - LessWrong经典英语资料.pdfVIP

LW LOGIN The ‘ petertodd’ phenomenon 158 by mwatkins 46 min read 15th Apr 2023 48 comments Glitch Tokens Eldritch Analogies GPT Language Models AI Frontpage Note: Nothing in this post is intended to vilify or defame any actual human named Peter Todd. I recognise that the existence of this phenomenon may be an uncomfortable fact for anyone with that name and do not wish to cause anyone discomfort. However, I feel that this phenomenon cannot be safely ignored, if we consider the role that powerful language models are rapidly assuming within global human civilisation. I will always refer to ‘ petertodd’ (with leading space and speech marks) to make clear that I am referring to a GPT token rather than a person. Readers should be advised that this post contains some (GPT-3-generated) obscene, violent and otherwise disturbing language. Many thanks to Jessica Rumbelow, janus, Cleo Nardo, Joe Collman, plex and Eliezer Yudkowsky for helpful discussions and comments on earlier drafts. Introduction is will be the rst in a sequence of posts, containing little analysis or speculation about the phenomenon in question and simply presented as an initial report of some unexplained ndings. It’s important to note that the ‘ petertodd’ phenomenon varies between GPT models[1], so it may be better thought of as a family of closely related phenomena. ere are models not discussed here with their own avours of ‘ petertodd’-triggered behaviour (see this Twitter feed for examples in code- davinci-002). is post is just a rst attempt to sketch some of the contours of the overall shape of the thing. All boldface prompts (and references to prompts) below link out to JSON les containing, typically, 250 completions at the stated temperature, plus temperature 0 completions, controls produced with random