Open Models Almanac
Methodspost-trainingsupervised fine-tuning

single mention · post-training

Freeze the vision encoder during pretraining and unfreeze for SFT

Freezes the vision encoder during pretraining, then unfreezes it for SFT.

The sources on this page name it without explaining how it works. The line above is the extraction's one-line gloss. See supervised fine-tuning for the methods filed beside it.

Filed under post-training › supervised fine-tuning