Let me just copy the final section before we then all dive into the supporting evidence.

Takeaways

What we take from this project is not that one 27B model now answers questions that it wouldn't before. It is that the bias layer in a Chinese-aligned model is shallow. It is applied late, it is applied thin, and fifty million tokens of carefully chosen supervision removes it without disturbing anything underneath. The values in any open model are only ever one fine-tune away. The users of a model deserve optionality in which values they work with, and everyone, including researchers, users and the American public, deserves high quality, open-weight models whose values can be objectively tested and adjusted.