mrq bd0bc10ec0 added LoRA policy to decide what layer of the model gets adapted based on simple inclusion/exclusion terms		2024-06-17 13:05:06 -05:00
..
arch
__init__.py
ar_nar.py	ugh	2024-06-15 12:29:03 -05:00
base.py	residual_in_fp32=False for mamba arch backends because it breaks the classifier (output projection / lm head / what-have-you) under AMP	2024-06-15 12:08:03 -05:00
experimental.py
lora.py	added LoRA policy to decide what layer of the model gets adapted based on simple inclusion/exclusion terms	2024-06-17 13:05:06 -05:00
nar.py