Skip to content

Issues: PKU-YuanGroup/MoE-LLaVA

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

61 Open 33 Closed
61 Open 33 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

How to use bf16 on V100
#97 opened Dec 12, 2024 by caichaoxiang
AssertionError: The model has moe layers, but None of the param groups are marked as MoE. Create a param group with 'moe' key set to True before creating optimizer
#96 opened Dec 11, 2024 by Asunatan
Lora + MoE
#95 opened Oct 13, 2024 by Paul0113Jin
[Usage] Can not download vision tower with directly 'sh scripts/v1/qwen/finetune_moe.sh'.
#94 opened Oct 11, 2024 by Routin
[Question] 论文table 7的 non-MoE LLaVA-phi的train scripts和eval scripts
#92 opened Sep 16, 2024 by sharkdrop
[Question] Is there any Moe checkpoint of Qwen1.5 or Qwen2 released?
#91 opened Aug 28, 2024 by double-fire-0
[Question] How to eval textqa
#90 opened Aug 21, 2024 by fanminshi
1
[Question] Step 3 loss curve
#89 opened Aug 16, 2024 by fanminshi
1
[Question] Question about the tokenizer of required pretrained model stabilityai/stablelm-2-1_6
#88 opened Aug 5, 2024 by Taylorfire
1
[Question] In paper Table 6, why variant (d) is better than variant (c)?
#87 opened Aug 1, 2024 by pkumc
[Feature request] 是否會訓練更進一步的模型
#86 opened Jul 17, 2024 by gesen2egee
Training of Stage 3 , 第三阶段训练，代码中实际的训练参数与论文不符
#85 opened Jul 11, 2024 by Wuyingwen
1
[Question] What exactly does the language model mean?
#84 opened Jun 27, 2024 by dana-niu
[Discussion] What is the expert relationship between different layers with the same index? If not, what is the role of figures 4, 5 and 6 in the paper?
#83 opened Jun 5, 2024 by meteorlium
[Question] ValueError: Unknown image tower: /hy-tmp/LLaVA/clip-vit-large-patch14-336
#82 opened May 25, 2024 by FanshuoZeng
5
Can the confidence coefficient of an answer be obtained?
#81 opened May 24, 2024 by IsabelJimenez99
[Question] Inconsistency on MoE Layer Number in paper and model config
#80 opened May 10, 2024 by QAQdev
[Usage] ADD windows support for more exposure
#79 opened May 9, 2024 by mr-lab
can you please give me python script to use API with your demo ?
#78 opened May 8, 2024 by gamesubzero
Moe finetuning error
#77 opened May 7, 2024 by sahilqure
[Question] 多图collate_fn
#76 opened May 7, 2024 by PangziZhang523
[Question] 能解释一下llava_arch中的class LlavaQWenMetaForCausalLM(LlavaMetaForCausalLM)这个类吗
#74 opened Apr 28, 2024 by 20191864218
[Feature request] Support Llama3
#71 opened Apr 21, 2024 by xiweideng
[Question] About parameter ep_size
#70 opened Apr 19, 2024 by puppy2000
[Usage] tokenizer.pad_token_id == None？
#69 opened Apr 18, 2024 by sjtu-cz
1
ProTip! Mix and match filters to narrow down what you’re looking for.