790 Commits

Author SHA1 Message Date
hiyouga
7bfbcb1fe3 vllm + lora support
Former-commit-id: 8cb86ba355195f5d6dcb95ee6b6b7203463a34db
2024-04-25 20:24:31 +08:00
hiyouga
6be321b5da fix #3374
Former-commit-id: 0097d7968b3b570e1705caff26f42d9ed71ad974
2024-04-25 19:56:49 +08:00
hiyouga
9aeb88c426 add export_device in webui #3333
Former-commit-id: 30ebd3652809d73941e0a5e4a8be11d989faf98d
2024-04-25 19:02:32 +08:00
hiyouga
36be12a3b7 update tool template
Former-commit-id: c72a1981859818c257c5271d32e03c9d3c344206
2024-04-25 00:21:34 +08:00
hiyouga
21fac4c98c fix log level
Former-commit-id: 8d21302f6201b3f33c10f61f3559bd95be3363c2
2024-04-24 23:42:59 +08:00
hiyouga
83404c4fa9 support new special token #3420
Former-commit-id: f5c6a47f5193ab3a6c137580992bdcce0b31fdd5
2024-04-24 23:39:31 +08:00
hoshi-hiyouga
12f852b8d4 fix phi template
Former-commit-id: 14a1ff665eaebfc618229efbe96f09848d52faec
2024-04-24 13:55:14 +08:00
hoshi-hiyouga
a88873116a fix webchatmodel
Former-commit-id: dc6d8b5dc42c363dd180aaf90c9a2f2d0cce6725
2024-04-24 13:54:21 +08:00
hoshi-hiyouga
7cfcd69c64 fix inference in llamaboard
Former-commit-id: 5e631915157083b61e2d5a183e0c91f2d11f416e
2024-04-24 13:53:39 +08:00
hiyouga
a5eabbe933 add olmo 1.7
Former-commit-id: 86a3fb3a141d2702b15af08df36ffcf9b3d6de14
2024-04-24 05:50:50 +08:00
hiyouga
aa25716a5d add dbrx and jamba models
Former-commit-id: ce35c80b4b00152185285d6064939803d14487f0
2024-04-24 05:39:52 +08:00
hiyouga
94c8219575 fix bug
Former-commit-id: 38e164fe4aaea6f0baf121a720291ca42643ba8c
2024-04-24 05:21:18 +08:00
hiyouga
ad24a2a0c9 fix bug
Former-commit-id: 271c24d2c82d645fa9072e6de94ca38f20411537
2024-04-24 05:10:07 +08:00
hiyouga
c05027d14a remove redundant code
Former-commit-id: 4a7a7ad2bcdc493458084f5f3d384239228b7d5a
2024-04-24 05:02:18 +08:00
hiyouga
5420905a2e support unsloth generate
Former-commit-id: 0ef1ad9f505dba71db9342f524cc3a7565e5e09e
2024-04-24 04:46:53 +08:00
hiyouga
03f2e3284a refactor patcher
Former-commit-id: 263cfe1294f5c3188f5e8d65791f35ee0d87315a
2024-04-24 03:02:23 +08:00
hiyouga
d2bb1b3a6b reenable sdpa and fast tok by default
Former-commit-id: 9e00902dbedc71d55743d1bf237843506a557891
2024-04-24 02:18:44 +08:00
hiyouga
35c4a2c212 fix #3347 #3387
Former-commit-id: c253c18185a29b59190f3e0ed236c2bb4c788085
2024-04-24 01:30:16 +08:00
hiyouga
1e4010a1fb support phi-3
Former-commit-id: 7e8ffa9beee3893e051ceeade443bd56c4a07b1c
2024-04-24 00:28:53 +08:00
hiyouga
1d2e372a8e update readme
Former-commit-id: d4eaee262a64e716ce475dc4eb18d8d9697d8dd8
2024-04-22 17:09:17 +08:00
hiyouga
348f29aa50 set dev version
Former-commit-id: b9557887d7506ff57b2b2bf490092aac4e4becf0
2024-04-21 23:14:30 +08:00
hiyouga
c8fe3f544b release v0.6.3
Former-commit-id: 947572af8de201669598f54735f35b50bb719d71
2024-04-21 23:13:23 +08:00
hiyouga
0f1ad7140f fix #3366
Former-commit-id: dc20237455c36de44f8922539d7dfadd8bedb12f
2024-04-21 21:34:25 +08:00
hiyouga
233e167f68 fix optimizers
Former-commit-id: f811eee2fa12a89a55a9c5d3a05a1521b4347727
2024-04-21 20:40:54 +08:00
hiyouga
1d341dcd83 fix #3365
Former-commit-id: 415ce41e8fa887e980e5bd575c8e95bd4076b90b
2024-04-21 19:20:18 +08:00
hiyouga
d16561e7a4 fix bug in galore optimizer
Former-commit-id: c05ac23261a5a8ba893c2918a43dc7777307407b
2024-04-21 18:53:22 +08:00
hiyouga
f8e219dc81 fix mod stuff
Former-commit-id: cf3988226e6398c67bb2955578e436fc505aa5c5
2024-04-21 18:11:10 +08:00
hoshi-hiyouga
3365cc8cf0 Merge pull request #3338 from astramind-ai/main
Adding Mixture of Depth

Former-commit-id: 4da2ece53353b63e672ff529d6beba41ff710c14
2024-04-21 18:05:52 +08:00
hoshi-hiyouga
3a5e68b7d9 fix #3348
Former-commit-id: aa5e921c00f60074eceb2f9d4d8837cc713edba6
2024-04-20 10:34:09 +08:00
hiyouga
b3b5b530d1 fix #3352
Former-commit-id: f315f8e8ec916b82bac94a159e55839ff155c6b5
2024-04-19 22:40:01 +08:00
hiyouga
9225c15c88 fix llama3 template
Former-commit-id: 20e95250168fbe081c779b2e1ff23f5df3ce02f7
2024-04-19 15:46:51 +08:00
Marco
abd9fed445 fix small typo
Former-commit-id: 5638a03cd0cf8119ff366b3b3e303b5a2351b065
2024-04-18 20:33:29 +02:00
Marco
44cda2eece Added Mixture of Depths
Former-commit-id: 75dd98b9abc847e22cb263c17ebcd2ca5dd98345
2024-04-18 20:31:24 +02:00
hoshi-hiyouga
8397808d1d support llama3
Former-commit-id: c1eabb751a5fd73b710714451b146732e0ed4558
2024-04-19 01:13:50 +08:00
hiyouga
9e1bd6420d fix #3324
Former-commit-id: 5e710c4ac331f3400534d33b2646c4108c898d98
2024-04-18 15:34:45 +08:00
hiyouga
619264c854 tiny fix
Former-commit-id: 86399ca8c06273c42c2b184664ae25d3405b3bf6
2024-04-18 00:22:17 +08:00
hiyouga
ce9bdb3509 add mixtral 8x22B models
Former-commit-id: eccbeecff0909e1fa124b5439ffbbfbc5607e1d6
2024-04-17 23:35:59 +08:00
hiyouga
0c8d6369ac add CodeQwen models
Former-commit-id: 9f6094241391f8f717818c8ba94e11d1791b4a5c
2024-04-17 23:27:22 +08:00
hiyouga
bee796f6b5 fix #3316
Former-commit-id: 7395e9e90a209228ff563ab54319955608850fc3
2024-04-17 22:54:34 +08:00
hiyouga
9f6349a333 fix #3317
Former-commit-id: 7dce1763be4374cf616d96db95ae964ff510a9d6
2024-04-17 22:17:19 +08:00
hiyouga
171a029c5e lint
Former-commit-id: 917d65ce65024d17a5030bc57083a427cfae16d7
2024-04-16 18:21:09 +08:00
hoshi-hiyouga
eaefaa0fe0 Merge pull request #3291 from codemayq/main
support for previewing custom dataset in directory format

Former-commit-id: 40d89152282101a7c08f53e72c2ad7124a0595f3
2024-04-16 18:12:09 +08:00
hiyouga
d301f0a64b Update parser.py
Former-commit-id: 92c2133896c20054db86dd53508c982e39bd5ca0
2024-04-16 18:09:31 +08:00
hiyouga
0a1578e4e3 update readme and gradio version
Former-commit-id: 4029b60ddcbd15b5354503c51178f0f5e7e9aedf
2024-04-16 18:09:16 +08:00
hiyouga
a4167fd925 support badam for all stages
Former-commit-id: 7a1380646119bfe6855f73dd90570defcea05281
2024-04-16 17:44:48 +08:00
hoshi-hiyouga
42084e08ae Merge pull request #3287 from Ledzy/badam
[Feature] Add BAdam algorithm

Former-commit-id: 10a5e1e65b34b03e5ca2a41bf6ded09a3fb25f0c
2024-04-16 17:32:16 +08:00
hoshi-hiyouga
9d23f5dc89 Update utils.py
Former-commit-id: 01147536b2bb507e87e033fa696e9eb39fe96bbe
2024-04-16 17:30:12 +08:00
hoshi-hiyouga
5978427ae0 Update trainer.py
Former-commit-id: c6163be1444c00dd000f288e2f834968bd932981
2024-04-16 17:29:52 +08:00
hoshi-hiyouga
c7c216069c Update utils.py
Former-commit-id: 7edf4dbed88b8034282f14fd6e0cb6f7f9e5f805
2024-04-16 17:29:30 +08:00
hoshi-hiyouga
cde9d1b917 Update patcher.py
Former-commit-id: 494e6a1e05b38f5ff61d83327303614f53c92e64
2024-04-16 17:29:19 +08:00