Supervised fine-tuningTaking a trained model and training it a bit more on your own examples so it gets better at one specific job.Full definition → introduced a blackmail propensity in every model tested, and what survived it was constitutional content merely present during midtraining — its structure mattered less.
Terms in this piece · Glossary
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
alignment — The work of making AI systems actually pursue what their builders and users intend, rather than something subtly or dangerously different.
articleMisalignment Has a Personality: A Big Five Account of Emergent MisalignmentHasibur Rahman, Smit Desai
articleAlignment Forecasting: Predicting Misalignment From Training DataChen Yueh-Han, Bruce W. Lee, Ilia Sucholutsky, Tomek Korbak