SFT on model specs … Does this improve generalization?
I would expect having the constitution memorized to help, but not as much as practice in then applying it correctly. So I would SFT overtrain (or even distill, if I had a suitable teacher) on the constitution, but I would also SFT train on discussions applying it to specific situations, such as CoT traces — marginal ones where the discussion is non-trivial seem most interesting. Again, this requires a better teacher.
I would expect having the constitution memorized to help, but not as much as practice in then applying it correctly. So I would SFT overtrain (or even distill, if I had a suitable teacher) on the constitution, but I would also SFT train on discussions applying it to specific situations, such as CoT traces — marginal ones where the discussion is non-trivial seem most interesting. Again, this requires a better teacher.