Joe Foran

Results 3 issues of Joe Foran

I'm using qlora on a machine with 4 32GB V100 gpus. If I use only 2 of the GPUs, training proceeds without any problem but when I use all 4...

I am trying to connect to a juypter lab instance running on a remote server. The version of jupyter lab is 4.0.2 and my version of ein is 20230622.1757 I...

In the algorithm outlined in the original paper, the threshold for whether adapted momentum is applied or not is set to ρt > 4, however looking at the code the...