Fix adapter v2 llm.int8 inference#323

Draft

diormiu wants to merge 8 commits intoLightning-AI:mainfrom

diormiu:fix-adapterv2-int8

Commits on May 24, 2023

convert adapterv2 weight to input dtype
diormiu
committed

Commits on May 25, 2023

dequantize adapter/layers weights
diormiu
committed

Commits on May 26, 2023

Update adapter_v2.py
rasbt
authored

Commits on Jun 2, 2023

Commits on Jun 7, 2023

add model.reset_cache() back
rasbt
authored