Skip to content

Issues: Qihoo360/360-LLaMA-Factory

Beta
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

14 Open 22 Closed
14 Open 22 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Support for Sequence Parallelism (SP) in the Multimodal Pipeline?
#56 opened May 29, 2025 by Eisenhower
1
对于sp， 当用ulysses替换ring_flash_attn时,为什么loss很大
#55 opened May 13, 2025 by cangshuli
关于华为NPU的支持
#47 opened Apr 17, 2025 by githubwqj
1 task done
2
SP does NOT work with liger kernel good first issue Good for newcomers
#36 opened Mar 22, 2025 by XD-BDIV-NLP
1 task done
3
360-LLAMA-Factory框架能否实现长序列的训练加速？还是说只是会减少显存占用？
#30 opened Mar 15, 2025 by StarDewXXX
1
Support for transformers versions < 4.46.1 will soon be discontinued due to issues with incorrect gradient accumulation.
#29 opened Mar 14, 2025 by chenzihan1
1 task done
1
When selecting data to pack, could an option be added to allow for random pack instead of using the default greedy pack?
#27 opened Mar 14, 2025 by cizhenshi
Invalid shape error when using Zero3
#20 opened Mar 11, 2025 by shiningliang
1 task done
1
训练到第一个checkpoint，运行验证数据集的时候报错：TypeError: iteration over a 0-d tensor
#15 opened Feb 25, 2025 by Shiquan0304
1
datasets.Dataset.map too slow even with num_proc
#10 opened Feb 2, 2025 by luckyfan-cs
7
Any plan to support long-sequence DPO for audio models? enhancement New feature or request
#6 opened Jan 17, 2025 by qiuyedyc
@HaoshengZou
1
ArrowInvalid: Column 4 named images expected length 360 but got length 352 enhancement New feature or request
#5 opened Jan 16, 2025 by DhruvaBansal00
1 task done
@HaoshengZou
7
请问该框架支持预训练的序列并行吗？ enhancement New feature or request
#4 opened Jan 16, 2025 by YihengLiu1996
1 task done
@HaoshengZou
2
Unable to Use Validation (valid) Feature While Starting Sequence Parallelism 无法使用验证（valid）功能 bug Something isn't working good first issue Good for newcomers
#2 opened Jan 14, 2025 by wuyuhan3z
1 task done
2
ProTip! Exclude everything labeled bug with -label:bug.