 Artem Chumachenko
|
1c89c5c7ff
fix rebase
|
2 年之前 |
 justheuristic
|
aa5e7e350c
Deep prompt inference (#65)
|
2 年之前 |
|
Artem Chumachenko
|
01c1e198b8
Fix merge conficts
|
2 年之前 |
|
Artem Chumachenko
|
88e6a75996
Add part of deepprompts
|
2 年之前 |
|
Artem Chumachenko
|
c003830cc6
fix
|
3 年之前 |
|
Artem Chumachenko
|
1afd59a071
introduce hypo_ids
|
3 年之前 |
|
Artem Chumachenko
|
f62c65ec23
fixes
|
3 年之前 |
|
Artem Chumachenko
|
53e19de6e0
Add tests
|
3 年之前 |
|
Artem Chumachenko
|
ade986ca58
Return multibatch mode
|
3 年之前 |
 Alexander Borzunov
|
54ad745bed
Warn that current instructions involve 6B model but we will replace them soon (#63)
|
3 年之前 |
|
Alexander Borzunov
|
5f0c5329d4
Update readme with arxiv link and more discussions (#62)
|
3 年之前 |
|
Alexander Borzunov
|
9bea7b9ea8
Update bullet points with feedback from Tim and other people (#61)
|
3 年之前 |
|
Alexander Borzunov
|
7653562aa1
Use latest version of Petals scheme, shrink Petals logo (#59)
|
3 年之前 |
|
Alexander Borzunov
|
2eb5843852
Update readme for the 1st public release (#57)
|
3 年之前 |
 Pavel Samygin
|
0be21775af
remove transformer block, implement as sequential of size 1 (#54)
|
3 年之前 |
|
Artem Chumachenko
|
77220c718c
Add shallow prefix-tuned inference (#55)
|
3 年之前 |
|
justheuristic
|
d271b75dd4
Let users specify sequence length instead of assuming 2048 (#52)
|
3 年之前 |
 Dmitry Baranchuk
|
948877149c
Fix recovering for sequential_backward (#50)
|
3 年之前 |
|
Dmitry Baranchuk
|
24ba3433e4
[Fix] make distributed seq cls to not create the full bloom model (#49)
|
3 年之前 |
|
justheuristic
|
f12d0deee9
[quickfix 1/n] remove expensive assertions in inference code (#48)
|
3 年之前 |
|
Dmitry Baranchuk
|
0fd2caa4be
Convert actual model weights (#46)
|
3 年之前 |
|
justheuristic
|
a2634001e9
Reduce vocabulary size in test model, fix bug in routing when overlapped (#45)
|
3 年之前 |
|
Dmitry Baranchuk
|
5745882c67
fix rpc_forward_stream
|
3 年之前 |
|
Dmitry Baranchuk
|
6095f58681
Deep distributed prompt tuning (#42)
|
3 年之前 |
|
justheuristic
|
9460220a10
make pytest outputs more verbose (#44)
|
3 年之前 |
|
Dmitry Baranchuk
|
c4aa1f49df
Rename 350m -> 560m (#43)
|
3 年之前 |
|
Artem Chumachenko
|
79a9ff2b2e
introduce hypo_ids
|
3 年之前 |
|
Dmitry Baranchuk
|
11a424837f
integrate mixed-8bit model (#39)
|
3 年之前 |
|
Alexander Borzunov
|
7d39d46966
Use "PETALS" as the readme title (#40)
|
3 年之前 |
|
Dmitry Baranchuk
|
04a2b6f5e3
Support various backend dtypes & async serialization (#38)
|
3 年之前 |
