mbzuai-oryx LLaVA-pp issues

mbzuai-oryx / LLaVA-pp

🔥🔥 LLaVA++: Extending LLaVA with Phi-3 and LLaMA-3 (LLaVA LLaMA-3, LLaVA Phi-3)

741 stars 49 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

ValueError: Trying to set a tensor of shape torch.Size([128257, 4096]) in "weight" (which has shape torch.Size([128256, 4096])), this look incorrect.

#31 basteran opened 1 month ago
1
How much memory are needed for fine-tuning llava-llama3

#30 simplelifetime opened 1 month ago
0
Finetuning with lora output never ends.

#29 gyupro opened 1 month ago
4
Total Parameters: 0

#28 believewhat opened 1 month ago
0
convert_to_hf

#27 Jayantverma2 opened 1 month ago
0
Saved weights during tuning with LoRA

#26 fangruizhu opened 1 month ago
0
【Error when merging LoRA weights】

#25 Luo-Z13 opened 1 month ago
3
finetune error about model size

#24 Skylight-Lark closed 1 month ago
5
inference error

#23 tsw123678 closed 1 month ago
1
'PreTrainedTokenizerFast' object has no attribute 'legacy'

#22 tsw123678 closed 1 month ago
3
could you please support deepseek v2？

#21 thesby opened 2 months ago
0
what is s2 in vision load tower and how it effects training

#20 Jayantverma2 opened 2 months ago
1
Training Issue

#19 DevonPeroutky opened 2 months ago
1
Can use Lora+base model. but for merging Lora+base is error

#18 hellangleZ opened 2 months ago
4
Tokenization mismatch in Phi-3 when finetune process

#17 hellangleZ closed 2 months ago
5
Update README.md

#16 jbn opened 2 months ago
0
Llava 1.6?

#15 ElliottDyson opened 2 months ago
3
Tokenization mismatch in LLaVA-LLaMA-3

#14 Luo-Z13 closed 2 months ago
2
Exactly same as our study! Interesting!

#13 mu-cai closed 2 months ago
2
Installation Issue

#12 orrzohar opened 2 months ago
2
zero3.json is missing

#11 matbee-eth closed 2 months ago
1
Can't load the HF model

#10 vfragoso closed 2 months ago
1
S2 finetuning

#9 xmu-xiaoma666 closed 2 months ago
1
Train with llava-llama3

#8 hellangleZ closed 2 months ago
9
Using phi-3 and LLava but some fields of phi3 network not support

#7 hellangleZ closed 2 months ago
21
inference score with ScienceQA

#6 xmu-xiaoma666 closed 2 months ago
0
what is the learning rate then finetuning all LLM parameters during SFT stage?

#5 xmu-xiaoma666 closed 2 months ago
2
app

#4 chandlergis closed 2 months ago
1
Vídeos support?

#3 RaulKite closed 2 months ago
1
Clarification of ScienceQA score

#2 Isaachhh closed 2 months ago
2
inference examples

#1 SkalskiP opened 2 months ago
12