Skip to content

Issues: state-spaces/mamba

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

299 Open 172 Closed
299 Open 172 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

Optimizing the bwd pass of Mamba 2
#530 opened Aug 18, 2024 by Hprairie
2
Gradient explosion in Mamba2 training, norm and loss divergence
#529 opened Aug 16, 2024 by edwko
1
Results vary greatly across experiments
#528 opened Aug 16, 2024 by William-HYWu
Vanishing gradient problem with more layer
#527 opened Aug 15, 2024 by yjdy
Is it okay to put S6(MAMBA) and gated MLP blocks like a transformer? (Also please open the discussion tab)
#526 opened Aug 15, 2024 by qnixsynapse
How can I avoid using causal-conv1d?
#525 opened Aug 14, 2024 by CacatuaAlan
11
Some questions about the shape of A,B,C,D
#524 opened Aug 13, 2024 by Aristo23333
4
clarification on how to interpret kernel size for conv1d
#523 opened Aug 13, 2024 by ScottHoang
1
Exploding gradients if ngroups is higher than 1.
#522 opened Aug 12, 2024 by DanFosing
3
Is it possible to get ssm_states at specific seq len instead of just the last state?
#518 opened Aug 9, 2024 by Lily-Le
1
mamba_split_conv1d_scan_combined error
#516 opened Aug 6, 2024 by hokiyoshi
1
Mamba for encoder-decoder
#515 opened Aug 6, 2024 by simple-boy
2
Changing expansion_factor resulted in 'c10::DistBackendError
#514 opened Aug 5, 2024 by ScottHoang
1
How to use sequence parallelism in Mamba2?
#512 opened Aug 3, 2024 by andregraubner
Implement a "crossmamba" block for different length sequences
#511 opened Aug 2, 2024 by YihangChen9
3
Replace mamba1 with mamba2 and training becomes very slow!
#510 opened Aug 1, 2024 by YQ-097
2
How to calculate Flops for mamba_chunk_scan_combined and mamba_split_conv1d_scan_combined function
#509 opened Aug 1, 2024 by dumpmemory
1
RuntimeError: CUDA error: unknown error
#508 opened Aug 1, 2024 by JoyceMind
big scan dim cause cuda error
#505 opened Jul 31, 2024 by mmm-cc
3
CUDA error when using Mamba2 with long context
#503 opened Jul 30, 2024 by titzehong
2
The output of forward and step are different
#500 opened Jul 29, 2024 by yjdy
Size mismatchs when loading mamba2-130m
#499 opened Jul 29, 2024 by chadyu9
7
error installing "pip install mamba-ssm"
#497 opened Jul 28, 2024 by achintyasarkar
5
Some deprications for torch 2.4.0 and triton 3.0.0
#496 opened Jul 28, 2024 by Anri-Lombard
1
why GPU memory cost of Mamba2 Block > full self-attention block? And how to reduce this memory cost when training?
#495 opened Jul 26, 2024 by thucz
1
ProTip! Add no:assignee to see everything that’s not assigned.