-
Notifications
You must be signed in to change notification settings - Fork 25
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
[bug] DeepSeekV4/V4-Flash 在 PP>1 + FP8 下训练时导出 lora adapter 时报错: _set_o_group_proj_grouped 未处理 mg_attn=None
bugSomething isn't workingSomething isn't workingStatus: Open.#148 In modelscope/mcore-bridge;与NVIDIA自己的 Megatron Bridge 相比, 这个项目速度快吗
questionFurther information is requestedFurther information is requestedStatus: Open.#145 In modelscope/mcore-bridge;GLM5.2 微调调用dsa的时候,不开重计算参数不对齐,开启重计算是正常的
bugSomething isn't workingSomething isn't workingStatus: Open.#143 In modelscope/mcore-bridge;AttributeError when linear_decoupled_in_proj is enabled
bugSomething isn't workingSomething isn't workingStatus: Open.#110 In modelscope/mcore-bridge;训练qwen3.5moe系列时 swift的 moe_grouped_gemm需要为true
enhancementNew feature or requestNew feature or requestStatus: Open.#38 In modelscope/mcore-bridge;dsa indexer TP
questionFurther information is requestedFurther information is requestedStatus: Open.#36 In modelscope/mcore-bridge;