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

255 Open 967 Closed
255 Open 967 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]: the training procedure exit without any error logs or stuck in some steps using multi-nodes with booster api bug Something isn't working
#4353 opened Jul 31, 2023 by zhangvia
[BUG]: stuck when scaling to 8 nodes with 8 GPUs per node bug Something isn't working
#4351 opened Jul 30, 2023 by Atopos-309
[BUG]: 在 interleaved pipeline schedule 情况下，模型训练 loss 不下降 bug Something isn't working
#4348 opened Jul 28, 2023 by huangting4201
1
[BUG]: llama training script cannot run in branch of example/llama bug Something isn't working
#4344 opened Jul 28, 2023 by Hap-Zhang
[BUG]: Training is slower than DeepSpeed in llama 7B? bug Something isn't working
#4342 opened Jul 27, 2023 by Hap-Zhang
[BUG]: llama example 3d parallel OOM even with cpu offload. bug Something isn't working
#4339 opened Jul 27, 2023 by uygnef
[shardformer] support sequence parallel
#4335 opened Jul 26, 2023 by FoolPlayer
@FoolPlayer
[BUG]: ModuleNotFoundError: No module named 'colossalai._C.cpu_adam' bug Something isn't working
#4327 opened Jul 25, 2023 by Aaricis
[BUG]: can not run batch2_seq2048_flash_attn for 65b llama bug Something isn't working
#4326 opened Jul 25, 2023 by uygnef
3
[flashattn] Support Flash Attention to v2.0.1
#4322 opened Jul 25, 2023 by kkk55596
1
[BUG]: Data parallelism and model parallelism have completely different weights bug Something isn't working
#4319 opened Jul 25, 2023 by bobo0810
12
[BUG]: Multi-rank on same device bug Something isn't working
#4318 opened Jul 25, 2023 by Gy-Lu
1
[BUG]: sft training 13B model outputs 75go model bug Something isn't working
#4308 opened Jul 23, 2023 by allaccs
1
[BUG]: TypeError: 'LazyTensor' object is not callable bug Something isn't working
#4296 opened Jul 20, 2023 by upwindflys
[plugin] add 3d parallel plugin enhancement New feature or request
#4294 opened Jul 20, 2023 by ver217
1
@ver217
[BUG]: 使用RedPajama-Data-1T-Sample数据集训练llama-7b模型报错 bug Something isn't working
#4289 opened Jul 20, 2023 by Maxhyl
2
[shardformer] support pipeline for ChatGLM
#4286 opened Jul 19, 2023 by CjhHa1
@CjhHa1
[shardformer] support GLM/ChatGLM
#4282 opened Jul 19, 2023 by klhhhhh
1
@klhhhhh
[FEATURE]: Support for LLaMA-2 enhancement New feature or request
#4281 opened Jul 19, 2023 by lvcc2018
3
[BUG]: ATTENTION! tensor.type_as() will return None with colossalai bug Something isn't working
#4280 opened Jul 19, 2023 by densechen
[shardformer] add tensor type of attn mask in Coloattention
#4262 opened Jul 18, 2023 by flybird1111
@flybird1111
bloom support jit fused operator
#4235 opened Jul 14, 2023 by flybird1111
@flybird1111
[zero] optimize the step time
#4220 opened Jul 12, 2023 by Gy-Lu
1
@Gy-Lu
[BUG]: dataloader num is not correct bug Something isn't working
#4207 opened Jul 10, 2023 by zhangvia
[FEATURE]: FP8 mixed precision training via Transformer Engine enhancement New feature or request
#4199 opened Jul 7, 2023 by sbhavani
ProTip! Exclude everything labeled bug with -label:bug.