Universal checkpoint for zero stage 1 #2284
Refactor universal checkpointing and tensor fragments
4b87f300
Merge branch 'master' into olruwase/refactor_universal_checkpoint
4317b846
Formatting
dfc816df
Merge branch 'master' into olruwase/refactor_universal_checkpoint
21aa55a9
Merge branch 'master' into olruwase/refactor_universal_checkpoint
c6838919
Merge branch 'master' into olruwase/refactor_universal_checkpoint
89df0b3c
Merge branch 'master' into olruwase/refactor_universal_checkpoint
622e7ab8
Merge branch 'master' into olruwase/refactor_universal_checkpoint
115fe422
Merge branch 'master' into olruwase/refactor_universal_checkpoint
3c09adaa
Merge branch 'master' into olruwase/refactor_universal_checkpoint
d40b9236
Support zero stage1; Expand TP dim
7cf02358
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
ece4ce37
Remove debug prints
cae21725
Merge branch 'olruwase/zero_1_2_universal_ckpt' of github.com:microso…
48b62c29
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
1ace0257
Detect sharded optimizer state
529f2d88
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
93246ece
Merge master
45320590
Format fixes
024baa8a
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
a2f592d3
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
697287d0
tjruwase
changed the title Universal checkpoint for zero stage 1 & 2 Universal checkpoint for zero stage 1 3 years ago
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
a5e99007
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
bfefdec6
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
b78c5f64
tjruwase
marked this pull request as draft 3 years ago
Encode reshaping guide
e3465292
Merge branch 'olruwase/zero_1_2_universal_ckpt' of github.com:microso…
1447fc23
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
1bebe2e9
tjruwase
marked this pull request as ready for review 3 years ago
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
a5afb811
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
bb85c69b
More symbolic constants
c929f89f
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
6ba5ad46
mrwyattii
approved these changes
on 2022-09-26
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
83ecf190
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
0f1738d6
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
6c6823fe
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
16d26b9c
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
a26458a8
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
48d291a7
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
e358ee6d
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
4f51d0a9
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
f30ad0fe
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
88bf0452
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
329cb826
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
0d9dc104
Merge branch 'master' into olruwase/zero_1_2_universal_ckpt
37485ae1
tjruwase
merged
799120e7
into master 3 years ago
mrwyattii
deleted the olruwase/zero_1_2_universal_ckpt branch 2 years ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub