datamllab LongLM issues

datamllab / LongLM

[ICML'24 Spotlight] LLM Maybe LongLM: Self-Extend LLM Context Window Without Tuning

https://arxiv.org/pdf/2401.01325.pdf

MIT License

548 stars 54 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Cohere command r

#42 flaviusburca opened 3 weeks ago
1
是否有示例代码支持对safetensors格式LLM启用SelfExtend

#41 WSC741606 closed 3 weeks ago
5
Differences with ReRoPE

#40 siyuanseever closed 1 month ago
1
Run example.py Error: Failed to modify the attention method of LlamaForCausalLM

#39 tuzeao-tal closed 1 month ago
2
Passkey retrieval (needle in a haystack)

#38 MarsJacobs closed 3 weeks ago
2
Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! when resuming training

#37 humza-sami opened 1 month ago
0
llama3 is not working.

#36 rayjang opened 2 months ago
1
Question about equation 4 and Table 5 caption in paper

#35 MarsJacobs closed 1 month ago
3
Questions regarding group query/key positional index

#34 MarsJacobs closed 2 months ago
2
Long context

#33 ssrisunt closed 2 months ago
0
Question | Has anyone tried this with GGUF models?

#32 Cognitus-Stuti opened 2 months ago
1
Something wrong in modify_method_of_instance function

#31 meizhen-nlp closed 2 months ago
2
What effect on qwen1.5 will be if i use self-extend trick?

#30 WeixuanXiong opened 3 months ago
3
FlashAttention does not work for Batch size > 1

#29 Mooler0410 opened 3 months ago
0
Add self extend flash attention v2 implementation core file

#28 qingquansong closed 3 months ago
0
OOM on LongBench

#27 YerongLi closed 3 months ago
1
System file cleanup

#26 grasskin closed 3 months ago
0
[Add] Gemma Example

#25 ariG23498 opened 3 months ago
0
Example for gemma & use with Ollama

#24 vishal-android-freak closed 3 months ago
5
llama_self_extend_patch_4_36 is not work

#23 YL-9 closed 3 months ago
7
Example for phi2?

#22 JoanZhou closed 3 months ago
7
Can it be implemented on qwen1.5?

#21 win10ogod closed 3 months ago
2
Flash Attention implementation is coming

#20 Mooler0410 closed 3 months ago
0
Export weights

#19 fakerybakery closed 3 months ago
1
Support for Phi2 / Mixformer

#18 anthony-chaudhary closed 4 months ago
1
Support with vLLM

#17 Aniketto16 opened 5 months ago
4
OOM when length is 16k

#16 dengxiaotian123 closed 4 months ago
4
没有报错，但是显示不出来实验结果

#15 guanzy2012 closed 3 months ago
2
运行代码后还是无法回答长文本

#14 Eren139 closed 4 months ago
4
How to reproduce the results on LongBench

#13 yihedeng9 closed 3 months ago
5
None error check

#12 leotam closed 5 months ago
0
Release the SOLAR model

#11 kungfu-eric closed 5 months ago
1
Long input series makes oom

#10 seanxuu closed 3 months ago
22
Requires excessive computing resources when inference

#9 zhhvvv closed 3 months ago
3
vllm integration

#8 Chirobocea closed 5 months ago
1
fix for mps

#7 HonorRong opened 5 months ago
0
fix for ops

#6 HonorRong closed 5 months ago
0
Update README to add how to integrate this repo to my existing code

#5 rohan-paul closed 5 months ago
0
Update README.md

#4 eltociear opened 5 months ago
0
add support for apple silicon

#3 HonorRong opened 5 months ago
0
Phi2 implementation and Suggestions

#2 agokrani closed 5 months ago
4
Flash Attention Support?

#1 0-hero closed 5 months ago
4