b6586

github-actions released this 25 Sep 20:15

835b2b9

model : add GroveMoE support (#15510)

* add GroveMoE support

* remove constexpr that fails on certain compilers

* revert crude scalar div implementation, use cast

* build_attn_inp_kv_unified -> build_attn_inp_kv

* fix build_attn

* re-apply ffn_exps regex changes

Assets 15