mrq edf1e66bf9 layerskip_r=6 fries the model so hard the loss is sub-1...		2024-11-01 17:06:07 -05:00
..
emb
engines	added option to load lora directly from the model file itself with --lora	2024-10-26 00:13:10 -05:00
ext
models	third time's the charm (for some reason it escaped me that I should treat early exit loss as an aux_loss to be used with the normal loss, as if I was training a MoE's router)	2024-11-01 12:50:37 -05:00
utils
__init__.py
__main__.py	added option to load lora directly from the model file itself with --lora	2024-10-26 00:13:10 -05:00
config.py	layerskip_r=6 fries the model so hard the loss is sub-1...	2024-11-01 17:06:07 -05:00
data.py
demo.py	layer skip training implemented (need to gut the inferencing from the repo, and to actually see if the model can benefit from this)	2024-10-30 20:05:45 -05:00
export.py
inference.py	added option to load lora directly from the model file itself with --lora	2024-10-26 00:13:10 -05:00
plot.py
samplers.py
train.py
webui.py	added option to load lora directly from the model file itself with --lora	2024-10-26 00:13:10 -05:00