Fork
0
代码
介绍
代码
Issues
Pull Requests
流水线
Actions
讨论
Wiki
项目成员
分析
项目设置
Fork
0
master
MindSpeed-LLM
/
mindspeed_llm
/
features_manager
/
transformer
/
flash_attention
下载当前目录
ascend-robot
[pytorch][feature] kvallgather supports TND
045b6da8
创建于
3月5日
历史提交
文件
最后提交记录
最后更新时间
alibi_feature.py
[pytorch][bugfix] baichaun2 no-fa-adapt Co-authored-by: jzh6229<jiangzhihui4@huawei.com>
7 个月前
fusion_attention_feature.py
[pytorch][bugfix]modify the default value of pre_tockens Co-authored-by: dingzicha1997<dingzilin@huawei.com> # message auto-generated for no-merge-commit merge:
!3589
merge master into master [pytorch][bugfix]modify the default value of pre_tockens Created-by: dingzicha1997 Commit-by: dingzicha1997 Merged-by: ascend-robot Description: [pytorch][bugfix]modify the default value of pre_tockens from 64K to 1M See merge request: Ascend/MindSpeed-LLM
!3589
6 个月前
reset_attention_mask_feature.py
[pytorch][feature] kvallgather supports TND Co-authored-by: Jia_Austin<dengjia6@huawei.com> # message auto-generated for no-merge-commit merge:
!4277
merge fix_te_tnd into master [pytorch][feature] kvallgather supports TND Created-by: Jia_Austin Commit-by: Jia_Austin Merged-by: ascend-robot Description: ## What this PR does / why we need it? feat: TE tnd ## Does this PR introduce any user-facing change? NA ## How was this patch tested? Turn on and off TE CP TND See merge request: Ascend/MindSpeed-LLM
!4277
2 个月前