Haoquan Zhou

icon indicating a website link https://seiunsky0131.github.io/

icon company @aliyun icon location Cloud Valley, Hangzhou icon location MLE @aliyun | MS Information @Umich @UMSI | B.E. ECE @SJTU

Results 1 comments of Haoquan Zhou

SFT FlashAttention questions

> What about the warning: > > You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU...