Two fascinating new papers on LLM interpretability from Anthropic
https://simonwillison.net/2025/Mar/27/tracing-the-thoughts-of-a-large-language-model/
If you have a fediverse account, you can quote this note from your own instance. Search https://fedi.simonwillison.net/users/simon/statuses/114236613585215146 on your instance and quote it. (Note that quoting is not supported in Mastodon.)