llama-lora-fine-tuning icon indicating copy to clipboard operation
llama-lora-fine-tuning copied to clipboard

Published 20 hours ago • verified publisher icon git-cloner

training stuck

Open mz2sj opened this issue 2 years ago • 1 comments

As the picture below,It didn't start training for a long time.Is it reasonable or the reason of V100 8-bit matmul is slow? Thanks for your help. image

mz2sj avatar Jun 16 '23 09:06 mz2sj

Nothing to do with that warning, on V100,load tokenizer take half an hour

little51 avatar Jun 17 '23 00:06 little51