Skip to content

[Models] Update SWA RoPE theta for MLA/GQA attention#8077

Merged
Jiang-Jia-Jun merged 3 commits into
PaddlePaddle:developfrom
chang-wenbin:mla_gqa_swa_rope_theta
Jun 26, 2026
Merged

[Models] Update SWA RoPE theta for MLA/GQA attention#8077
Jiang-Jia-Jun merged 3 commits into
PaddlePaddle:developfrom
chang-wenbin:mla_gqa_swa_rope_theta

Commits

Commits on Jun 25, 2026