Zhao (Bill) Xu

Results 2 comments of Zhao (Bill) Xu

> use `deepspeed --include "localhost:0,1,2,3"` in full_multi_gpu/single_node.sh Thanks for your reply! Is the setting in lora_multi_gpu/single_node.sh right if I need to specify the last 4 GPUs for lora sft? `CUDA_VISIBLE_DEVICES=4,5,6,7...

使用Chrome Dark Mode测试,发现没有原生支持,应该得使用额外插件?