 dbaranchuk
|
d1abb2f169
address the comments
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
ed86b36911
address the comments
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
b60eedc8ad
Merge remote-tracking branch 'origin/main' into efficient-forward-backward
|
%!s(int64=3) %!d(string=hai) anos |
 justheuristic
|
f0cffbf67e
Miscellaneous fixes to automatic tests (#35)
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
3cc4e0b9f0
update server deployment scripts
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
f976191551
efficient forward & backward
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
7de3acf909
Fix is_subsequence (#32)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
f0c7383181
Implement RemoteSequential slicing and extra repr, add tests (#30)
|
%!s(int64=3) %!d(string=hai) anos |
 Artem Chumachenko
|
6ee942e915
Add GenerationMixin class (#29)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
177d81bea6
CI: use GIT_REF_NAME instead of GIT_HEAD_REF (#28)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
49242d8100
[WIP] Fix CI runs in master (#27)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
01ed4db750
Fix default branch in CI (#26)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
ccdcefe405
Add instructions to test the full model (#25)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
eb0a6be716
Clean up readme (#24)
|
%!s(int64=3) %!d(string=hai) anos |
|
justheuristic
|
e2711a033b
Add automated tests (#23)
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
f5463812ad
Shallow prompt tuning (#22)
|
%!s(int64=3) %!d(string=hai) anos |
 Alexander Borzunov
|
7e9f337a63
Remove excess line from readme
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
db966a76dd
rm heuristic for num_handlers
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
f114a6d417
set default num_handlers=16
|
%!s(int64=3) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
f3cf5f4d8d
Fix choose_best_blocks()
|
%!s(int64=3) %!d(string=hai) anos |
|
Alexander Borzunov
|
75856e4769
Measure and cache network & compute throughput (#21)
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
ac7df18dfa
Merge pull request #19 from learning-at-home/lm_head
|
%!s(int64=3) %!d(string=hai) anos |
|
Dmitry Baranchuk
|
fd0bf064f3
minor refactoring
|
%!s(int64=3) %!d(string=hai) anos |
|
Alexander Borzunov
|
aba43f1308
Implement block selection on servers (#20)
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
21e1f42f04
mv set_requires_grad to remote_model
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
5168a3405a
fix comments
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
79280c4371
refactoring
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
6bffeff0a1
fix
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
b3cc9e0d99
add LM head for DistributedBloomCausalLM
|
%!s(int64=3) %!d(string=hai) anos |
|
dbaranchuk
|
df42822f26
LM head for CausalLM & chunked forward
|
%!s(int64=3) %!d(string=hai) anos |
