mrq c1fcd889d5 reverted automatically disabling split loss calc, since it seems that it's actually cacling loss on prom causes the oddities, maybe		2024-06-01 12:34:59 -05:00
..
__init__.py
distributed.py	ugh	2024-05-11 16:47:19 -05:00
sampler.py
trainer.py	reverted automatically disabling split loss calc, since it seems that it's actually cacling loss on prom causes the oddities, maybe	2024-06-01 12:34:59 -05:00
utils.py	nasty hotfix for transformer's Mixtral throwing an error when batch sizes > 1	2024-01-26 19:41:12 -06:00
wrapper.py	leverage between xformers and torch.backends.cuda.sdp_kernel for attention	2024-05-11 17:14:05 -05:00