issues
search
jongwooko
/
distillm
Official PyTorch implementation of DistiLLM: Towards Streamlined Distillation for Large Language Models (ICML 2024)
https://arxiv.org/abs/2402.03898
145
stars
21
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
fix: Handle different separator token values for uint16/uint32 dtypes
#14
jlamprou
opened
3 days ago
0
Problem during finetuning qwen2 models
#13
j-datta
opened
2 months ago
0
Dataset link seems invalid now
#12
aaab8b
closed
2 months ago
2
Remove unused imports in `/distillm/__init__.py`
#11
BM-K
closed
2 months ago
0
Download the training/evaluation data
#10
ypw-lbj
closed
4 months ago
2
gdk mismatch in equation?
#9
ogimgio
closed
4 months ago
0
test12
#8
sheldoer
closed
4 months ago
0
incorrect loss functions based on KLD definition
#7
mohammadreza-molapanah
closed
5 months ago
4
the code of task-specific experiments
#6
wangfan120
opened
6 months ago
0
Inconsistnet with your reported scores of teacher models
#5
liuxy1103
closed
8 months ago
12
OPTS+=" --kd-ratio 1.0"
#4
liuxy1103
closed
8 months ago
2
How to download the models after sft
#3
liuxy1103
closed
8 months ago
4
The metric of GPT-4 Eval
#2
liuxy1103
closed
8 months ago
1
Update README.md
#1
eltociear
closed
10 months ago
0