原书出处
Designing Large Language Model Applications
Data augmentation In this technique, the teacher is shown a set of seed input-output examples. Based on the seed examples, the teacher generates similar input-output examples, constituting the distillation dataset. Note that both the input and output are generated by the teacher model in this setting. The limitation of this technique is that the teacher is unable to generate sufficiently diverse examples. Intermediate representations This
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch09.html
EPUB SHA-256:1b4d2347fec2f949fcddfbaf4fad597a1ae20887a096cfae64c59a619b4781bf