transformers
33a8e68d - Add GlmMoeDsa (#43858)

Commit
113 days ago
Add GlmMoeDsa (#43858) * draft * for review only * update ignore layers * add config * update * rename * fallback * update * 1 * update * fix attention and date * remove pretraining_tp and improve tests * style * remove pretraining_tp * fix config * fix compile * remove wrong integration test * fix * better --------- Co-authored-by: zRzRzRzRzRzRzR <2448370773@qq.com>
Author
Parents
Loading