lwmlyy

icon indicating a website link https://scholar.google.com/citations?user=YyoelDMAAAAJ&hl=en

icon company Shanghai University of Finance and Economics icon location Shanghai icon location NLP, WSD, machine learning, deep learning

Results 1 issues of lwmlyy

CUDA OOM during ckpt saving for Llama2-70b

14 comment

Hi, I am using 8*a100-80gb to lora-finetune Llama2-70b, the training and evaluation during epoch-1 went well, but went OOM when saving the peft model. The nightly version of pytorch is...

triaged