Skip to content

Issues: linkedin/Liger-Kernel

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

37 Open 40 Closed
37 Open 40 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

Optional dependency on transformers
#258 opened Sep 19, 2024 by DuarteMRAlves
Loss does not drop when using Liger Kernel at Qwen2.5
#257 opened Sep 19, 2024 by Se-Hun
Are you even allowed to do these ops inplace?
#254 opened Sep 18, 2024 by mayank31398
1
reverse KL and JSD feature
#252 opened Sep 17, 2024 by yundai424
@yundai424
1
AutoLigerKernelForCausalLM.from_pretrained discards hub_kwargs_names
#250 opened Sep 16, 2024 by uris-opti
@shimizust
3
ValueError when Loading Qwen2-VL Model with Liger Kernel
#249 opened Sep 16, 2024 by rahatarinasir
1
Support for Cohere models
#248 opened Sep 15, 2024 by nyxkrage
1
Is not compatible with DoRA? bug Something isn't working
#245 opened Sep 11, 2024 by gotzmann
1
Reasons for upcasting the logits dtype outside the kernel
#241 opened Sep 10, 2024 by yzhangcs
7
Benchmarking phi3 on single A100 40gb GPU: unable to reproduce benchmark results
#236 opened Sep 9, 2024 by cosmicBboy
3
Triton error on AMD GPUs
#231 opened Sep 8, 2024 by eminorhan
8
Torch compiled FLCE is 2x faster than the current FLCE
#227 opened Sep 7, 2024 by ByronHsu
12
MoE kernel feature
#206 opened Sep 4, 2024 by ByronHsu
4
Z Loss in CE feature
#197 opened Sep 3, 2024 by Fr0do
2
@shivam15s
2
Lenient Test
#192 opened Sep 3, 2024 by jon-chuang
1
Improve the efficiency of the RMSNorm aggregation
#179 opened Aug 30, 2024 by lancerts
1
@S1ro1
5
Division by zero when total_n_ignore = 0 bug Something isn't working p0
#177 opened Aug 30, 2024 by 44670
@shivam15s
1
[fun] mamba.triton fun
#156 opened Aug 29, 2024 by ByronHsu
[feat] Add LitGPT Integration feature p0
#139 opened Aug 28, 2024 by shivam15s
Benchmark against Flash attention repo
#137 opened Aug 28, 2024 by zzhhjjj
@lancerts
[feat] Add support for encoder-only transformers (e.g. BERT) feature
#131 opened Aug 27, 2024 by OxxoCodes
[feat] support for DeepseekV2 feature help wanted Extra attention is needed huggingface
#129 opened Aug 27, 2024 by tmm1
4
[feat] FusedLinearCrossEntropy support for Gemma2 feature
#127 opened Aug 27, 2024 by yundai424
@troy1729
4
[AMD] Implement Flash Attention in Triton to enable transformers to run with Flash Attention on AMD GPUs. AMD feature
#126 opened Aug 27, 2024 by ByronHsu
3
[fun] llama.triton fun
#119 opened Aug 27, 2024 by ByronHsu
7
ProTip! Updated in the last three days: updated:>2024-09-16.