interpreting GPT: the logit lens — LessWrong

1 min read Original article ↗

x

interpreting GPT: the logit lens — LessWrong