OwnerSunshine530 9f73505caa perf: MoE top-k加权改scatter+mul+sum(在[E,N,D]上),省permute大clone+gather(profile clone 8%) ...

数学等价(top-k索引互异,scatter无冲突),零AUC风险。延续'减kernel'方向。
moe_fused_weight默认开,test_moe_dense_matches_loop已覆盖。

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-06-19 20:22:16 +08:00

..

code

perf: MoE top-k加权改scatter+mul+sum(在[E,N,D]上),省permute大clone+gather(profile clone 8%)

2026-06-19 20:22:16 +08:00

main.ipynb

chore: 更新 notebook 环境和文档

2026-06-12 20:51:14 +08:00

任务提交接口说明.md

chore: 初始化 CTI 推理优化项目

2026-06-03 13:49:30 +08:00