Skip to content

Issues: datawhalechina/unlock-deepseek

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

32 Open 14 Closed
32 Open 14 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Param "max_seq_length" in FastLanguageModel shouldn't be set to "max_completion_length"
#52 opened Mar 29, 2025 by Dielianss
docs 里边的md文件都没有内容
#51 opened Mar 12, 2025 by zyjy200
2
4090单卡跑代码报错：shape '[4, 997, 16, 0]' is invalid for input of size 8167424
#50 opened Mar 10, 2025 by 1066196847
1
CUDA version
#49 opened Mar 6, 2025 by Cbphcr
1
复现效果失败
#48 opened Mar 1, 2025 by a1exyu
3
我调整了一下训练的配置，显示我在cpu训练
#47 opened Feb 27, 2025 by wzk2239115
9
ImportError: cannot import name 'patch_compiled_autograd' from 'unsloth_zoo.patching_utils' (/miniconda3/envs/logic/lib/python3.9/site-packages/unsloth_zoo/patching_utils.py
#46 opened Feb 25, 2025 by GorgeousWang
3
请问unsloth方法使用的基座模型是什么？我用7b训崩了
#44 opened Feb 22, 2025 by dream-tentacle
9
list indices must be integers or slices, not str
#43 opened Feb 21, 2025 by zy1998-c
1
RuntimeError: The expanded size of the tensor (10) must match the existing size (11) at non-singleton dimension 0
#42 opened Feb 19, 2025 by Mrkkew
1
DeepSpeed版本训练报错RuntimeError: The size of tensor a (780) must match the size of tensor b (781)
#40 opened Feb 18, 2025 by tize-72
6
可否在单机多卡下运行 unsloth
#37 opened Feb 18, 2025 by hellobiek
1
单机单卡训练显存需求
#31 opened Feb 14, 2025 by robinjoe93
3
TypeError: Value.__init__() missing 1 required positional argument: 'dtype'
#30 opened Feb 14, 2025 by robinjoe93
5
启动训练失败问题
#28 opened Feb 12, 2025 by xiongbinbtbu
12
从某次推理开始全是！
#27 opened Feb 12, 2025 by momo4826
2
提示AttributeError: 'list' object has no attribute 'get'
#26 opened Feb 11, 2025 by Rvelamen
1
训练OOM
#22 opened Feb 10, 2025 by Mrkkew
5
训练时loss一直为0
#21 opened Feb 10, 2025 by FWLamb
10
DeepSeek-V3 对 MoE 的架构创新
#14 opened Jan 31, 2025 by anine09 MoE
DeepSeek-V2 对 MoE 的架构创新
#13 opened Jan 31, 2025 by anine09 MoE
DeepSeek-MoE 对 MoE 的架构创新
#12 opened Jan 31, 2025 by anine09 MoE
DeepSeek 系列模型中对 MoE 的改进
#11 opened Jan 31, 2025 by anine09 MoE
@anine09
DeepSeek Infra 工作概述
#10 opened Jan 31, 2025 by anine09 Keys
@Tangent-90C
PPO 与 GRPO 算法介绍
#9 opened Jan 31, 2025 by anine09 Reasoning Models
@linjh1118
ProTip! Exclude everything labeled bug with -label:bug.