0
代码
代码
Issues
Pull Requests
讨论
Wiki
分析
0
ascend-robotascend-robotcleancode
4bf77ad7创建于 3月30日历史提交
文件最后提交记录最后更新时间
triton
cleancode Co-authored-by: liyingxuan<liyingxuan3@huawei.com> # message auto-generated for no-merge-commit merge: !2323 merge master into 26.0.0 cleancode Created-by: liyx616 Commit-by: liyingxuan Merged-by: ascend-robot Description: ## What this PR does / why we need it? cleancode整改 ## Does this PR introduce any user-facing change? cleancode整改 ## 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-MM!23232 个月前
chunk_gated_delta_rule.py
[Bugfix]Counteract verl's autocast promotion (bf16 -> fp32) by restoring original dtype Co-authored-by: ruanhao566<ruanhao7@huawei.com> # message auto-generated for no-merge-commit merge: !2300 merge master into master [Bugfix]Counteract verl's autocast promotion (bf16 -> fp32) by restoring original dtype Created-by: ruanhao9656 Commit-by: ruanhao566 Merged-by: ascend-robot Description: ## What this PR does / why we need it? verl训练时会对计算autocast, 导致输出为fp32, 在triton计算时报dtype问题, 这里将dtype还原 ## Does this PR introduce any user-facing change? 不影响mm训练 ## 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-MM!23002 个月前
modeling_qwen3_5.py
[bugfix]修复RL训练时cp开启alltoall的shape不匹配的问题 Co-authored-by: Miss_min<qiaoxiaomin@huawei.com> # message auto-generated for no-merge-commit merge: !2298 merge master into master [bugfix]修复RL训练时cp开启alltoall的shape不匹配的问题 Created-by: Miss_min Commit-by: Miss_min Merged-by: ascend-robot Description: ## What this PR does / why we need it? Please describe the background and detailed changes of the PR. If it is a bugfix, please attach the related issue. ## 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-MM!22982 个月前