Skip to content

Issues: hpcaitech/ColossalAI

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

196 Open 321 Closed
196 Open 321 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
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

[BUG]: 使用 gemini，必须是2的幂的卡数，不然出现 assert chunk_size % self.pg_size == 0 bug Something isn't working
#2836 opened Feb 20, 2023 by 1024er
1
[BUG]: Error during training a reward model with lora_rank=16 bug Something isn't working
#2834 opened Feb 20, 2023 by haozhestat
[DOC]: There are a problem when I use multi-node documentation Improvements or additions to documentation
#2833 opened Feb 20, 2023 by zte-tcb
[BUG]: math.prod is not always available bug Something isn't working
#2832 opened Feb 20, 2023 by ErenChan
1
1
[PROPOSAL]: Publish the reproduction results on some mainstream tasks(both image and text). enhancement New feature or request
#2831 opened Feb 20, 2023 by flymark2010
1 task
1
[DOC]: Clear environment version scope documentation Improvements or additions to documentation
#2830 opened Feb 20, 2023 by binmakeswell
@FrankLeeeee
[FEATURE]: Optimization enhancement New feature or request
#2827 opened Feb 20, 2023 by programmin1
[BUG]: Title：Error when running train_reward_model.py with Facebook/OPT-350M model bug Something isn't working
#2825 opened Feb 19, 2023 by pariskang
1
[FEATURE]: Save model checkpoint with FP16 enhancement New feature or request
#2819 opened Feb 19, 2023 by larry-fuy
when Train the reward model, Get error
#2818 opened Feb 18, 2023 by zhangsanfeng86
[BUG]: Failed to run example bug Something isn't working
#2816 opened Feb 18, 2023 by SilenceGTX
1
[FEATURE] Patch meta information of operator.le() enhancement New feature or request
#2815 opened Feb 18, 2023 by Cypher30
1
@Cypher30
[FEATURE] Patch meta information of torch.finfo() enhancement New feature or request
#2814 opened Feb 18, 2023 by Cypher30
1
@Cypher30
[BUG]: I can not install colossalAI bug Something isn't working
#2811 opened Feb 18, 2023 by shendl1978
2
[BUG]: No module named 'chatgpt.nn' bug Something isn't working
#2810 opened Feb 18, 2023 by shendl1978
1
[FEATURE] Patch meta information of torch.where() enhancement New feature or request
#2800 opened Feb 17, 2023 by Cypher30
1
@Cypher30
[BUG]: ChatGPT Reward Model Eval Distance No Accumulated? bug Something isn't working
#2795 opened Feb 17, 2023 by Dcyx
1
4
[DOC]: Fix a small place in README.md under ChatGPT demo folder documentation Improvements or additions to documentation
#2794 opened Feb 17, 2023 by Fridge003
8
about chatGPT three steps documentation Improvements or additions to documentation
#2793 opened Feb 17, 2023 by leizhu1989
[FEATURE]: Patch meta information of torch.Tensor.size() enhancement New feature or request
#2780 opened Feb 17, 2023 by Cypher30
1
@Cypher30
[BUG]: run in Docker can not find parameter _forward_module.first_stage_model.encoder.conv_in.weight in the GeminiDDP module bug Something isn't working
#2778 opened Feb 17, 2023 by zhaoanbei
[BUG]: A100 24G，加载opt6.7b爆内存 bug Something isn't working
#2772 opened Feb 16, 2023 by iMountTai
2
[BUG]: Fail to load huggingface pretraining when use shardinit bug Something isn't working
#2770 opened Feb 16, 2023 by sega-hsj
[BUG]: Using mpirun to launch multi-node training stucks in colossalai.launch_from_openmpi bug Something isn't working
#2768 opened Feb 16, 2023 by pluiez
4
[BUG]: 用3张A100显卡使用examples的opt demo会爆显存 bug Something isn't working
#2758 opened Feb 16, 2023 by dreaaim
2
ProTip! Type g i on any issue or pull request to go back to the issue listing page.