Add JetMoE model #30005

ArthurZucker merged 122 commits into huggingface:main from main
yikangshen
yikangshen init jetmoe code
b63bcafb
yikangshen Merge branch 'huggingface:main' into main
03f646e8
yikangshen update archive maps
ed52b578
yikangshen remove flax import
150cd931
yikangshen fix import error
436a44cd
yikangshen update README
bcf597f6
yikangshen ruff fix
5c0400ec
yikangshen update readme
e61d131f
yikangshen fix
57b13ebc
yikangshen update config
1f27ad43
yikangshen fix issue
2ea5542f
yikangshen merge files
109a8c2f
yikangshen fix model bug
21a4c2d7
yikangshen fix test
9d542ac2
yikangshen auto fix
c5092b46
yikangshen model size
41f24364
yikangshen add comments
3052ce8d
yikangshen fix form
539cfb90
yikangshen add flash attention support
0f6af1d5
yikangshen fix attention head number
165e20d4
yikangshen fix init
68633f9b
yikangshen fix support list
d39a0e9e
yikangshen sort auto mapping
ef62bf3f
yikangshen fix test
c0a3076a
yikangshen fix docs
4d79ce65
yikangshen update test
e5336b50
yikangshen fix test
67aedd15
yikangshen fix test
c87de945
yikangshen change variable name
2f02e7e9
yikangshen fix config
fc39dcc2
yikangshen fix init
b4b57381
yikangshen update format
c370377d
yikangshen clean code
f443c293
yikangshen fix config
852ef61b
yikangshen fix config
9517a2b8
yikangshen change default config
30b826db
yikangshen update config
a18a67a6
yikangshen yikangshen marked this pull request as draft 1 year ago
yikangshen yikangshen marked this pull request as ready for review 1 year ago
yikangshen yikangshen changed the title Add JetMoE Add JetMoE model 1 year ago
gante gante requested a review from gante gante 1 year ago
fakerybakery
gante
gante commented on 2024-04-05
gante
yikangshen fix issues
88991b59
yikangshen Merge branch 'main' into main
7d23d95a
yikangshen update formate
28ed7c43
yikangshen update config argument
913dc9e6
yikangshen update format
5611f096
yikangshen
ArthurZucker
ArthurZucker
ArthurZucker commented on 2024-04-17
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
430ea6c6
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
3aff0697
yikangshen Merge branch 'main' into main
5528995b
yikangshen change to mixtral aux loss
c51e9870
yikangshen change to cache_position
4927e607
yikangshen debug
cc89ea38
ArthurZucker
yikangshen
yikangshen fix bugs
4dcbd268
yikangshen Merge branch 'main' into main
7a379707
yikangshen debug
5cdc9c74
yikangshen fix format
535c24c3
yikangshen fix format
192202e9
yikangshen fix copy
ddb11d07
yikangshen fix format
f9877f28
yikangshen fix format
7ec96dcf
yikangshen fix sort
2a0e123e
yikangshen fix sort
cecb26fc
yikangshen fix sort
5f2cffb4
yikangshen add copy comment
797a89b8
yikangshen add copy from
a7a6e2d5
yikangshen remove debug code
b37bb87a
yikangshen Merge branch 'main' into main
675867d7
yikangshen revert readme update
ae19e2cd
yikangshen add copy
d4787969
yikangshen debug
b8cdc4a6
yikangshen remove debug code
15170eef
yikangshen fix flash attention
6e464179
yikangshen add comments
be88983f
yikangshen
ArthurZucker
yikangshen Merge branch 'huggingface:main' into main
06a5d624
younesbelkada
younesbelkada commented on 2024-04-26
yikangshen
yikangshen clean code
8633fc55
yikangshen Merge branch 'main' of https://github.com/yikangshen/transformers
97125e3e
yikangshen clean format
d3002f3a
yikangshen fix format
bcad4fa4
yikangshen fix format
16f6fd88
yikangshen
HuggingFaceDocBuilderDev
younesbelkada
younesbelkada commented on 2024-04-30
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
71f6431f
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
fb26a0ef
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
6cb4df00
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
330a89ba
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
0ff62a91
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
f6ffe33d
yikangshen change variable name
e40e1702
yikangshen add copied from
7db942f6
yikangshen Merge branch 'huggingface:main' into main
8a2b5934
yikangshen fix variable name
05635bfd
yikangshen Merge branch 'main' of https://github.com/yikangshen/transformers
89142638
yikangshen remove deprecated functinos
ea7daa16
yikangshen sync to llama implementation
22a03f04
yikangshen fix format
303942af
yikangshen fix copy
30031242
yikangshen fix format
5b7101a3
yikangshen update format
5f335c31
yikangshen
younesbelkada
younesbelkada approved these changes on 2024-05-02
younesbelkada younesbelkada requested a review from ArthurZucker ArthurZucker 1 year ago
younesbelkada younesbelkada requested a review from amyeroberts amyeroberts 1 year ago
yikangshen remove repr
d5a66045
yikangshen add comment for moe weight
82069a1b
yikangshen Merge branch 'huggingface:main' into main
01733bd2
yikangshen fix copy
c6e5e8b5
ArthurZucker
ArthurZucker
ArthurZucker commented on 2024-05-08
yikangshen Update src/transformers/models/jetmoe/configuration_jetmoe.py
5cfc652a
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
42c02b04
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
976b4cf4
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
6588db55
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
dea51cba
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
410882aa
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
077e46a6
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
193a9efb
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
14512fc4
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
0bbfc87d
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
ecb03378
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
7f6d5296
yikangshen add comments and reformat config
4b327ba3
yikangshen fix format
7f447519
ArthurZucker
yikangshen fix format
6c0ea955
yikangshen Merge branch 'main' into main
a9e2c22c
yikangshen fix format
9c8081de
yikangshen update test
cf172045
yikangshen update doc string in config
41d1a703
yikangshen
ArthurZucker
ArthurZucker
ArthurZucker approved these changes on 2024-05-13
yikangshen Update src/transformers/models/jetmoe/modeling_jetmoe.py
58e56274
yikangshen update config doc
8341eeab
yikangshen Merge branch 'main' of https://github.com/yikangshen/transformers
71a29392
yikangshen update attention cache
9e8b7594
yikangshen Merge branch 'huggingface:main' into main
5c21dfee
yikangshen fix format
1b8ed083
yikangshen fix copy
060af341
ArthurZucker
yikangshen
ArthurZucker ArthurZucker merged ccdabc56 into main 1 year ago
ArthurZucker
yikangshen

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone