Skip to content

Issues: InternLM/xtuner

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

172 Open 337 Closed
172 Open 337 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

选择四卡训练卡住
#934 opened Sep 20, 2024 by AlittlePIE
intern2.5-20B微调 后词表长度不一致
#933 opened Sep 14, 2024 by topology1
使用lengthgroupedsampler代替原本的sampler后卡死
#932 opened Sep 13, 2024 by xcy9614
如何进行val和test？
#929 opened Sep 12, 2024 by Diyigelieren
version `GLIBCXX_3.4.29' not found
#928 opened Sep 11, 2024 by amannier
奖励模型问题
#926 opened Sep 10, 2024 by Eren139
transformers == 4.44.2 xtuner == 0.1.23 训练 qwen2 时报错
#925 opened Sep 7, 2024 by thomZ1
1
多机多卡训练报错ss1.ss_family == ss2.ss_family. 2 vs 10
#924 opened Sep 6, 2024 by sph116
请问与 llamaFactory 的训练 TGS 对比时的具体实验条件
#923 opened Sep 6, 2024 by shihanmax
报错Cannot find reference 'VarlenAttnArgsToMessageHubHook' in 'init.py'
#922 opened Sep 4, 2024 by hutiehua-1
1
有个疑问，计算Loss的时候并不是以reward_token_id最终loss计算的，为什么推理的时候可以以reward_token_id为准呢？
#921 opened Sep 1, 2024 by woshixiaobai2019
6
QwenVL支持
#920 opened Aug 31, 2024 by liyan1997
整合Liger Kernel: 最高效的Triton Training Kernels
#919 opened Aug 29, 2024 by ByronHsu
一些关于步数统计的疑问
#918 opened Aug 29, 2024 by young-chao
reward model训练完如何预测？
#916 opened Aug 28, 2024 by tcxia
1
qlora微调的模型是不支持中断后继续训练吗？
#915 opened Aug 27, 2024 by deep-practice
2
InternVL构造单图多轮对话数据的时候，每轮对话都需要加上<image>标签吗？
#913 opened Aug 27, 2024 by deep-practice
1
【应当修改哪个环境变量？】Setting ds_accelerator to cuda (auto detect) df: "/home/guochenchen/.triton/autotune": 没有那个文件或目录
#912 opened Aug 24, 2024 by gwoksansan
如何修改 master port
#911 opened Aug 23, 2024 by AislantVentus
train time decrease from 13 hours to 9
#910 opened Aug 23, 2024 by mylesgoose
LLaVa phi-3 sft 报错 ConnectionResetError: [Errno 104] Connection reset by peer
#909 opened Aug 22, 2024 by Yu-Yang-Li
微调数据集策略（dataset make confuse）
#908 opened Aug 22, 2024 by EasonQYS
internvl微调的数据集一条有多个jsonl文件和多个图片该怎么写config
#907 opened Aug 21, 2024 by mspythontu
How to modify the vision encoder of llava-llama3-8b?
#904 opened Aug 19, 2024 by Jason8Kang
fine-tuning codegeex4
#903 opened Aug 18, 2024 by sgjohnson1981
ProTip! Adding no:label will show everything without a label.