InternEvo is an open-sourced lightweight training framework aims to support model pre-training without the need for extensive dependencies.
310
stars
52
forks
source link
fix(mlp.py): fix mlp w1w2w3 init order to w1w3w2 #360
Open
huangting4201 opened 3 weeks ago
基于模型的执行顺序来调整模型的初始化顺序，这样才能使得isp overlap时prefetch module weight的顺序是对的