Open Models Almanac
Methodspost-trainingsupervised fine-tuning

single mention · post-training

Prompt warm-up with the pretrained backbone frozen

Warms up new soft prompts while keeping pretrained model weights frozen.

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 Prompt warm-up.

Filed under post-training › supervised fine-tuning