Open Models Almanac
Methodspost-trainingsupervised fine-tuning

single mention · post-training

Capability-targeted second-stage supervised fine-tuning

Adds a smaller SFT stage with a changed data blend to improve capabilities underlearned in the first stage.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See supervised fine-tuning for the methods filed beside it.

Also called small second stage of SFT.

Filed under post-training › supervised fine-tuning