xiaoachen98 Open-LLaVA-NeXT issues - Githubissues

xiaoachen98 / Open-LLaVA-NeXT

An open-source implementation for training LLaVA-NeXT.

234 stars 9 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Pretraining with im_start_end token

#18 Ali2500 opened 2 weeks ago
0
Is this still active?

#17 chris-hoertnagl closed 1 month ago
1
Inverse Loss Spike issue

#16 LaBaZh closed 1 month ago
1
Flashattention issue

#15 LaBaZh closed 1 month ago
1
srun: unrecognized option '--quotatype=reserved'

#14 LaBaZh closed 1 month ago
1
Would you plan to adapt it to qwen2-7B?

#13 Nastu-Ho opened 1 month ago
3
loss curve of llava-next-llama3

#12 simplelifetime opened 2 months ago
2
llama3 finetune time

#11 Xiaohui9607 closed 2 months ago
1
Vision Tower

#10 homiec closed 2 months ago
1
loss curve of SFT on vicuna-7b

#9 Xiaohui9607 opened 2 months ago
6
anyres in open-llava-next v.s. s2 in llava

#8 LIO-H-ZEN opened 2 months ago
0
About the MMMU performance

#7 LIO-H-ZEN opened 2 months ago
0
Is the param "self.image_newline" only tuned in SFT?

#6 huangrh99 opened 3 months ago
1
vit not saved

#5 RifleZhang closed 3 months ago
2
data generation code

#4 trinhvg closed 3 months ago
1
does it support llava-next-video yet?

#3 dragen1860 closed 3 months ago
1
[Question] About finetuning projector

#2 JY-CCK closed 3 months ago
1
Newbranch

#1 xiaoachen98 closed 3 months ago
0