 justheuristic
|
a707eaebc6
review
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
d94549eda8
review
|
%!s(int64=3) %!d(string=hai) anos |
 Aleksandr Borzunov
|
4f9ec08537
review
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
a9af1b4cf5
Merge branch 'fix-cache' of github.com:bigscience-workshop/petals into fix-cache
|
%!s(int64=3) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
0f5c427969
use humanfriendly
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
b470052581
add test for exceeding max length
|
%!s(int64=3) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
33f6bdfac2
review
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
9987fe83ba
Update src/server/handler.py
|
%!s(int64=3) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
f018e64df1
review
|
%!s(int64=3) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
612100ba3e
review
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
d6053a2516
forward metadata in sequential inference session
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
004aae3bf7
black-isort
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
0b09b33d07
fix cache size args, check it in tests
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
1d59e0c7be
Merge branch 'main' into fix-cache
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
2110b2cdc8
fix humane parser
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
11492c7097
clarify assert
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
d179ec138d
black-isort
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
a8d9159db6
enforce max_length
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
3d56314987
parse human-readable cache_size_bytes
|
%!s(int64=3) %!d(string=hai) anos |
 Dmitry Baranchuk
|
948877149c
Fix recovering for sequential_backward (#50)
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
24ba3433e4
[Fix] make distributed seq cls to not create the full bloom model (#49)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
f12d0deee9
[quickfix 1/n] remove expensive assertions in inference code (#48)
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
0fd2caa4be
Convert actual model weights (#46)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
a2634001e9
Reduce vocabulary size in test model, fix bug in routing when overlapped (#45)
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
5745882c67
fix rpc_forward_stream
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
6095f58681
Deep distributed prompt tuning (#42)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
9460220a10
make pytest outputs more verbose (#44)
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
c4aa1f49df
Rename 350m -> 560m (#43)
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
11a424837f
integrate mixed-8bit model (#39)
|
%!s(int64=3) %!d(string=hai) anos |
 Alexander Borzunov
|
7d39d46966
Use "PETALS" as the readme title (#40)
|
%!s(int64=3) %!d(string=hai) anos |
