zyxxmu/DSnoT

Official Pytorch Implementation of Our Paper Accepted at ICLR 2024-- Dynamic Sparse No Training: Training-Free Fine-tuning for Sparse LLMs

Python

Issues

Problem with reproducing results in Table1 on LLaMA2-7B
#5 opened 9 months ago by WailordHe
0
How to evaluate the zero-shot performance?
#4 opened 10 months ago by liuxiaozhu01
6
WrappedGPT object has no attribute 'H'
#3 opened a year ago by jonas-klesen
2
sparse model
#2 opened a year ago by kevinintel
2
Difference in default values and README example values
#1 opened a year ago by jonas-klesen
1