Support Qwen2-7b MLP in int4 and transpose_value_cache=True#11968

Merged

plusbang merged 3 commits intointel-analytics:mainfrom yangw1234:qwenint4

Sep 2, 2024

+49-11

Commits on Sep 2, 2024

Support Qwen2-7b mlp in int4

yangw1234
authored and
rnwang04
committed
fix style

yangw1234
authored and
rnwang04
committed
rm verify size
plusbang
committed