原书出处
Designing Large Language Model Applications
Learning schedules Toward the end of the training process, it is a good idea to lower the learning rate because you do not want to overshoot when you are so close to convergence. In a similar vein, you would like to prevent your model from learning too much from the first few batches of examples. In either case, we would like to be able to
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/ch06.html
EPUB SHA-256:1b4d2347fec2f949fcddfbaf4fad597a1ae20887a096cfae64c59a619b4781bf