yuhuixu1993 qa-lora issues

yuhuixu1993 / qa-lora

Official PyTorch implementation of QA-LoRA

MIT License

113 stars 10 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

is that right? could you tell me how to fix this error?

#36 LiZhangMing opened 4 months ago
2
Merging problem

#35 samuelqy opened 4 months ago
6
Hi can you tell me how to fix the error?

#34 LiZhangMing closed 4 months ago
0
Thanks for your helpful project,could you give us model checkpoint as shown in figure

#33 LiZhangMing opened 4 months ago
2
AWQ+LoRA available?

#32 RanchiZhao opened 5 months ago
0
After merge.py, is the model int4 data type?

#31 RanchiZhao closed 5 months ago
2
the difference between QA-Lora and simple GPTQ+LoRA in training

#30 RanchiZhao closed 5 months ago
0
ValueError: Target modules [] not found in the base model.

#29 akkkb opened 6 months ago
2
cannot reproduce the MMLU accuracy claimed in paper, could you release the script?

#28 wenjingk-xilinx opened 6 months ago
1
3bit not supported by autogptq with triton?

#27 wenjingk-xilinx opened 6 months ago
0
Discrepancy in Reproduced Results for LLaMA Tuning on Alpaca Dataset

#26 xxw11 closed 6 months ago
0
Request for Replication Script for LLaMA 7B on MMLU

#25 xxw11 closed 6 months ago
2
Encounter data type problem in train

#24 wenjingk-xilinx closed 6 months ago
3
encounter bug in "auto_gptq/modeling/_base.py"

#23 wenjingk-xilinx closed 8 months ago
2
About the question of derivation of merge_with_quantization in the paper

#22 LuletterSoul opened 9 months ago
1
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cuda:3! (when checking argument for argument mat2 in method wrapper_CUDA_mm)

#21 orangezfj opened 9 months ago
1
Finetuning Vision Foundation Models like OWL-ViT & Grounding Dino is possible? Any reference is available?

#20 solomonmanuelraj opened 9 months ago
0
Llama2 supported？

#19 skykiseki closed 6 months ago
7
How to set batchsize

#18 StiphyJay opened 10 months ago
1
Training with multi gpus, increase the batch size, and how to evaluate?

#17 shawnricecake opened 10 months ago
5
The loss cannot converge when finetuning Llama2-7b-GPTQ on 4090

#16 cyita opened 10 months ago
11
merge err: scales and qzeros dimension mismatch

#15 sscheng216 opened 10 months ago
4
How to support FLAN v2 dataset.

#14 ChenMnZ opened 10 months ago
0
quantize_config.json file

#13 thi3nnq opened 12 months ago
1
Adapter dimensions question

#12 sparverius opened 1 year ago
4
fixing lint issues

#11 sparverius closed 1 year ago
0
RuntimeError: self and mat2 must have the same dtype

#10 M-Elfeki opened 1 year ago
6
Equation, algorithm and experimental results question

#9 MarsJacobs closed 8 months ago
3
Is it possible to use an open source model that Huggingface has quantified？

#8 wangxiaochun520 opened 1 year ago
1
(Enhancement) Suggestion to incorporate GPTQ adapter merging into axolotl library

#7 jaredquekjz opened 1 year ago
1
Test

#6 sparverius closed 1 year ago
0
The paper is using Sum Pooling but the script is using Average Pooling

#5 fahadh4ilyas opened 1 year ago
1
Add missing parenthesis in peft_utils.py

#4 Trapper4888 closed 1 year ago
1
Can this be merged with normal 16bits model

#3 Kimiko-AI closed 1 year ago
1
Fix typo in README.md

#2 eltociear closed 1 year ago
0
What is the expected time for release of the code?

#1 Njasa2k closed 1 year ago
1