-
Notifications
You must be signed in to change notification settings - Fork 4.5k
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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
[BUG]: Lora load error
bug
Something isn't working
#6221
opened Feb 25, 2025 by
447428054
2 tasks done
[BUG]: EP16 negative split
bug
Something isn't working
#6220
opened Feb 25, 2025 by
447428054
2 tasks done
【Question】What is the minimum number of GPUs required to train deepseek 671B with GRPO? How about using LoRA?
#6219
opened Feb 25, 2025 by
LiuShixing
[BUG]: /bin/bash: line 0: export: `NPU-VISIBLE-DEVICES=0,1,2,3,4,5,6,7': not a valid identifier
bug
Something isn't working
#6217
opened Feb 24, 2025 by
Gera001
2 tasks done
Respecting regulations and stabilizing the ecosystem by activists
bug
Something isn't working
#6216
opened Feb 24, 2025 by
MASIHMIRSALI
2 tasks done
[BUG]: Precision overflow occurs when moe forward is performed
bug
Something isn't working
#6212
opened Feb 21, 2025 by
zh2333
2 tasks done
[BUG]: failed to install coati in npu docker environment
bug
Something isn't working
#6209
opened Feb 20, 2025 by
wangyuan249
2 tasks done
[BUG]: 该如何安装colossal到NPU上,看项目有相关描述,但没找到相关教程
bug
Something isn't working
#6205
opened Feb 20, 2025 by
obj12
2 tasks done
[DOC]: Update the documentation of ShardConfig for 1D, 2D, 2.5D, 3D tensor parallelism
documentation
Improvements or additions to documentation
#6197
opened Feb 18, 2025 by
giriprasad51
[FEATURE]: Expert Parallel for qwen/deepseek
enhancement
New feature or request
#6180
opened Jan 12, 2025 by
Guodanding
[BUG]: RuntimeError: mat1 and mat2 must have the same dtype, but got Float and BFloat16
bug
Something isn't working
#6169
opened Dec 25, 2024 by
balcklive
1 task done
[BUG]: Gemini saved an additional portion of the weights while using tie_word_embeddings=True
bug
Something isn't working
#6160
opened Dec 13, 2024 by
ericxsun
1 task done
[FEATURE]: Lora/QLora in GeminiPlugin and TorchFSDP
enhancement
New feature or request
#6138
opened Nov 16, 2024 by
ericxsun
[FEATURE]: support google/gemma-2-2b for Tensor Parallelism
enhancement
New feature or request
#6120
opened Nov 9, 2024 by
jing-4369
2
[BUG]: why duplicate PID appears on rank 0
bug
Something isn't working
#6111
opened Nov 3, 2024 by
ericxsun
1 task done
[BUG]: Llama3.1 HybridParallelPlugin train failed when pp_size>1
bug
Something isn't working
#6110
opened Nov 2, 2024 by
cingtiye
1 task done
[PROPOSAL]: FP8 with block-wise amax
enhancement
New feature or request
#6105
opened Oct 28, 2024 by
Edenzzzz
1 task
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.