1234 位用户此时在线
24小时点击排行 Top 10：
- 本站自动实时分享网络热点
- 24小时实时更新
- 所有言论不代表本站态度
- 欢迎对信息踊跃评论评分
- 评分越高，信息越新，排列越靠前
2
1
0
A new method was able to delete 40% of LLM layers with no drop in accuracy.
This makes them mich cheaper and faster.
The method combines pruning, quantization and PEFT.
They tested this across various open source models.
Each family of models had a maximum amount of layers…
时政
(
twitter.com
)
由
Lior⚡
提交
A new method was able to delete 40% of LLM layers with no drop in accuracy.
This makes them mich cheaper and faster.
The method combines pruning, quantization and PEFT.
They tested this across various open source models.
Each family of models had a maximum amount of layers that could be deleted before accuracy dropped:
- Mistral - 30%
- Lama 70B - 40%
- Llama 13B - 50%
Great paper.
点击图片查看原图
1
2
1
1
2
2
1
1
3
2
1
1
4
2
1
1
5
2
1
1
6
2
1
1
7
2
1
1
8
4
3
3
9
2
1
1
10
2
1
1
11
2
1
1
12
2
1
1
13
2
1
1
14
2
1
1
15
BREAKING: Supreme Court leaves injunctions against the Biden administration’s new Title IX rules in place.
2
1
1
BREAKING: Supreme Court leaves injunctions against the Biden administration’s new Title IX rules in place.
All nine Justices agree the rule imposing a new, speech-restrictive definition of sexual harassment should remain paused.
FIRE has consistently warned that the Department
时政
(
twitter.com)
•
FIRE
16
2
1
1
17
2
1
1
18
2
1
1
19
3
2
2
20
2
1
1
21
2
1
1
22
2
1
1
23
2
1
1
24
2
1
1
25
An amazing way to help individuals realize their worth and overcome shyness.
2
1
1
An amazing way to help individuals realize their worth and overcome shyness.
The shyest kid in class was asked to pick a slip - if it was blank, the whole class would be excused from homework.
The teacher had secretly ensured all the slips were blank, just to make the kid a
有趣
(
twitter.com)
00:00:26