Tied Crosscoders: Tracing How Chat LLM Behavior Emerges from Base Model

I had an idea for creating a crowdsourced database of AI prompts that no AI model could yet crack (wanted to use some of them as we're adding new models to Kilo Code).

I've seen a bunch of those

Article URL: https://goodenough.us/blog/2025-04-24-you-need-customers-to-succeed-in-small-business/

Article URL: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5106235

Comments URL:

Article URL: https://bigthink.com/starts-with-a-bang/21cm-magic-length/

Comments URL:

Article URL: https://rook2root.co/articles/20250424-manufacturing-consensus-on-x

Comments URL:

Article URL: https://www.ycombinator.com/companies/bild-ai/jobs/m2ilR5L-founding-engineer

Comments

Article URL: https://status.open-vsx.org/

Comments URL: https://news.ycombinator.com/item?