issues
search
veya2ztn
/
fast_retention
Speed up Parallel Retention about 2x times
2
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Does this implementation also support Multiscale Retention?
#5
Shreyas-Dongre
closed
1 year ago
4
How to use Parallel Retention?
#4
Shreyas-Dongre
closed
1 year ago
2
bfloat16 will get large error but speed up more
#3
veya2ztn
opened
1 year ago
0
Question about larger D
#2
syncdoth
opened
1 year ago
3
The head dimension can not larger than 32. $D>32$
#1
veya2ztn
closed
1 year ago
1