0
代码
代码
Issues
Pull Requests
讨论
Wiki
分析
0
ascend-robotascend-robotfeat(pytorch):ring cp support MLA/GQA
bd45114e创建于 4月25日历史提交
文件最后提交记录最后更新时间
megatron_data
feat(pytorch): Training in TND with GDN support Co-authored-by: mhh111<mahonghao1@huawei.com> # message auto-generated for no-merge-commit merge: !4354 merge 0327 into master feat(pytorch): Training in TND with GDN support Created-by: mhh111 Commit-by: mhh111 Merged-by: ascend-robot Description: feat(pytorch): Training in TND with GDN support See merge request: Ascend/MindSpeed-LLM!43541 个月前
processor
style(pytorch): SCA compliance rectification Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4355 merge master into master style(pytorch): SCA compliance rectification Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? SCA compliance rectification ## Does this PR introduce any user-facing change? no ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!43552 个月前
__init__.py
[pytorch][features]fsdp2 dataloader Co-authored-by: guihaowen666<guihaowen@huawei.com> # message auto-generated for no-merge-commit merge: !4041 merge br_master_fsdp2_dataloader_dev into master [pytorch][features]fsdp2 dataloader Created-by: guihaowen666 Commit-by: guihaowen666 Merged-by: ascend-robot Description: New dataloader support in fsdp2 backend 1 . The Converter is used to align tags in the dataset. 2. The template is used to align and encode data. 3. The processor is configured to process data into an input of the model. See merge request: Ascend/MindSpeed-LLM!40414 个月前
collator.py
style(pytorch): SCA compliance rectification Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4355 merge master into master style(pytorch): SCA compliance rectification Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? SCA compliance rectification ## Does this PR introduce any user-facing change? no ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!43552 个月前
converter.py
style(pytorch): SCA compliance rectification Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4355 merge master into master style(pytorch): SCA compliance rectification Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? SCA compliance rectification ## Does this PR introduce any user-facing change? no ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!43552 个月前
data_factory.py
feat(pytorch):ring cp support MLA/GQA Co-authored-by: mhh111<mahonghao1@huawei.com> # message auto-generated for no-merge-commit merge: !4412 merge ring0418 into master feat(pytorch):ring cp support MLA/GQA Created-by: mhh111 Commit-by: mhh111 Merged-by: ascend-robot Description: ## What this PR does / why we need it? ring cp support MLA/GQA ## Does this PR introduce any user-facing change? Please describe whether the PR will result in any user-facing usage changes. If there is related documentation, please specify its path. ## How was this patch tested? Please explain how to verify the correctness and effectiveness of this feature, as well as its usage constraints and limitations. See merge request: Ascend/MindSpeed-LLM!44121 个月前
data_utils.py
style(pytorch): SCA compliance rectification Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4355 merge master into master style(pytorch): SCA compliance rectification Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? SCA compliance rectification ## Does this PR introduce any user-facing change? no ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!43552 个月前
formatter.py
style(pytorch): SCA compliance rectification Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4355 merge master into master style(pytorch): SCA compliance rectification Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? SCA compliance rectification ## Does this PR introduce any user-facing change? no ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!43552 个月前
parser.py
style(pytorch): SCA compliance rectification Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4355 merge master into master style(pytorch): SCA compliance rectification Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? SCA compliance rectification ## Does this PR introduce any user-facing change? no ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!43552 个月前
template.py
[pytorch][doc]: fix template argument Co-authored-by: cjy840282<chenjingyi9@huawei.com> # message auto-generated for no-merge-commit merge: !4290 merge template into master [pytorch][doc]: fix template argument Created-by: cjy840282 Commit-by: cjy840282 Merged-by: ascend-robot Description: # Pull Request 模板 ---- ## What this PR does / why we need it? Fix template argument and register_template for empty. ## Does this PR introduce any user-facing change? Add explanations for the default and optional values of the template parameter in FSDP2. ## How was this patch tested? Only usability improvement. See merge request: Ascend/MindSpeed-LLM!42902 个月前
tokenizer.py
style(pytorch): SCA compliance rectification Co-authored-by: zhyebin01<zhangyebin@h-partners.com> # message auto-generated for no-merge-commit merge: !4355 merge master into master style(pytorch): SCA compliance rectification Created-by: zhyebin01 Commit-by: zhyebin01 Merged-by: ascend-robot Description: ## What this PR does / why we need it? SCA compliance rectification ## Does this PR introduce any user-facing change? no ## How was this patch tested? pipeline test passed See merge request: Ascend/MindSpeed-LLM!43552 个月前