GitLab

* add w4a16

* fix `deploy.py`

* add doc

* add w4a16 kernels

* fuse w1/w3 & bugfixes

* fix typo

* python

* guard sm75/80 features

* add missing header

* refactor

* qkvo bias

* update cost model

* fix lint

* update `deploy.py`

* fix build tests failure

* move src test cases to tests/csrc

* clean-up

* fix lint

* fix lint

* add GQA for llama2

* fix model conversion

* fix lint & remove dev log

* update news

* minor

* fix allocation size

* fix split_dim for w_qkv.bias

* format-11.1

* md-link-config

* feat(template): remote diff

* feat(cmake): use c++17

* build turbomind

* change namespace fastertransformer to turbomind

* change logger name