265 Commits

Author SHA1 Message Date
hiyouga
a73e7e022d add test mm plugin
Former-commit-id: ddea5cca5a3174de1dcc7fdee8ec69e77700b6bf
2024-08-31 01:53:38 +08:00
hiyouga
f389b6a676 tiny fix
Former-commit-id: 830511a6d0216da99520aee8b3a753d347a71fa9
2024-08-30 03:21:50 +08:00
hiyouga
228f745235 refactor mm training
Former-commit-id: 179c0558699e287cbf38a2d73bff47e86d589c5a
2024-08-30 02:14:31 +08:00
hoshi-hiyouga
0601cd69a9 fix bug
Former-commit-id: 365e6df71509569f59c40743c115f1a4b945ef0f
2024-08-30 02:05:26 +08:00
simonJJJ
cc4ea6a8aa update
Former-commit-id: a968a416d5e513320c97109229ca1e6ddc003cb1
2024-08-28 20:22:46 +08:00
simonJJJ
5e728ec221 initial-commit
Former-commit-id: b6a39847a10b417b09db4b5512dd835e9e4ce928
2024-08-28 16:51:35 +08:00
hoshi-hiyouga
a9e0274fae Merge pull request #5156 from YeQiuO/main
fix Llama-template's system prompt bug

Former-commit-id: 0b57175d3bd029675dae2f55995b7eeb4e9adc7a
2024-08-20 00:09:03 +08:00
hoshi-hiyouga
8f78d82e0b Update template.py
Former-commit-id: f5a075cb1c90f05bb0de26c6aea718f556c54623
2024-08-20 00:03:33 +08:00
hoshi-hiyouga
eb7a2557e7 Update template.py
Former-commit-id: c6822a217e1c296f4aedd9a2c7610acd1dbd443e
2024-08-19 23:40:16 +08:00
Huiyu Chen
649a5d55ec Add SailorLLM template
Former-commit-id: a594abe0321a718394a97b5a48ded16e2012c1f0
2024-08-15 15:10:14 +08:00
“Wzw”
db93661225 fix Llama-template's system prompt bug
Former-commit-id: 2e3eddcd0918b0c968ded0df7c82e3dcff870381
2024-08-12 19:22:12 +08:00
hiyouga
727193cdc9 follow #5115
Former-commit-id: 7d917e03e2df570139bae18227d9c7303a12de2a
2024-08-09 18:03:00 +08:00
hoshi-hiyouga
88036daf95 Merge pull request #5115 from YeQiuO/main
fix: `Train on the last turn only` truncate bug
Former-commit-id: 2c6dae45f7a7b72c961489ac407b1b444ab7752e
2024-08-09 17:58:27 +08:00
hoshi-hiyouga
7cbc5f3974 Update template.py
Former-commit-id: ae2a5221c109ae3474d219c37433be767abbee91
2024-08-09 16:27:42 +08:00
“Wzw”
a53c99ecda mask_history args verify valid
Former-commit-id: 2f8388b4f4195d934400ad9267d72e10ca4105a3
2024-08-08 10:12:01 +08:00
“Wzw”
cf3a209a93 fix mask_history tiny bug
Former-commit-id: cac07aac6196be026f723b2397a343d4fb675973
2024-08-08 10:09:33 +08:00
moontidef
d368f7b95d fix: fix the deepseekcoder template to avoid repeat problem
Former-commit-id: 56294831115f095135f72490a8a435434b2f0a11
2024-08-05 23:55:45 +08:00
hoshi-hiyouga
fc3443858d Merge pull request #4892 from piamo/main
update deepseek template

Former-commit-id: 3233efc8404972098665286d9dec7312dd6ecfab
2024-07-26 11:49:34 +08:00
hiyouga
1043118182 fix #4928
Former-commit-id: 6d557e8959678f9d4edbcb3d5a6dfba14b429b18
2024-07-24 17:00:29 +08:00
hiyouga
9aa1093362 fix flashattn + packing
Former-commit-id: 4adc6ce4abc718c25f39b316bfc3352d0d01ed1e
2024-07-21 17:07:45 +08:00
huangpan.foo
d72e3a2f01 update deepseek template
Former-commit-id: f5ca86ec95bb301df42ffaa6923fc3037a224e34
2024-07-19 15:02:54 +08:00
hiyouga
9573799224 follow #4878 fix #4684
Former-commit-id: 4715e5c5b8040b21e5f401f7e969b9fd2757d520
2024-07-18 22:06:12 +08:00
Shiyu Zhang
1538923eed 仅仅训练最后一轮对话
Former-commit-id: ab6198e4c099edeb1a400f58729cd617e8cd8e50
2024-07-18 15:30:25 +08:00
hoshi-hiyouga
07ea5796e5 Merge pull request #4691 from codemayq/feature-suppot-eval-dataset
add eval dataset support

Former-commit-id: 51eb379b44fad0336fc96c329ec98dc4528b9c2c
2024-07-15 01:00:34 +08:00
hoshi-hiyouga
3cf18efeef Update preprocess.py
Former-commit-id: da92f4a1b9c12a8e2489b964baba5e2c8e739ef1
2024-07-15 00:55:36 +08:00
hoshi-hiyouga
51265f81e7 Update parser.py
Former-commit-id: 145687997c86b8785e37dd60fbb9f3a5986730a6
2024-07-15 00:55:21 +08:00
hoshi-hiyouga
1e47c4a9d0 Update data_utils.py
Former-commit-id: 5c2a0e3b1d1afd2a9219d935d3421fffffc3a2c9
2024-07-15 00:54:34 +08:00
hoshi-hiyouga
f78cd9f9da Update loader.py
Former-commit-id: 860e3eb374947b72dcae88cab0a93ef561e3bfb3
2024-07-15 00:50:06 +08:00
hoshi-hiyouga
6978a9838f Update parser.py
Former-commit-id: b9760df588e64270a140d9111241c62c1cefe781
2024-07-14 23:04:34 +08:00
hiyouga
9cd850c3b9 fix gemma2 attention
Former-commit-id: aeafc68e169ae0ea5939cc81cb0cf89f0ca044b6
2024-07-13 23:33:45 +08:00
hiyouga
0a56404dd2 fix #4792
Former-commit-id: d7547d6b9e4c660897e3ce0f4022e08686c172d5
2024-07-13 22:07:58 +08:00
hiyouga
1fe104fd2c add codegeex4, internlm2.5
Former-commit-id: 349a5fbc934ac289cad44b4e3eb16f458b94710c
2024-07-06 16:16:47 +08:00
codingma
82e941ff61 1. add custom eval dataset support
2. merge load dataset and split dataset function


Former-commit-id: 963d97ba07e7efa3a4544c4d077283d9e112b3ad
2024-07-05 15:52:10 +08:00
hiyouga
a0fd90ce05 fix processors
Former-commit-id: 7215f3a8612b570cd322802d14db532927900117
2024-07-05 08:33:22 +08:00
hiyouga
a49956efd9 fix #4683
Former-commit-id: cbff0ea0db6971f8ced503a2f0cb6bc43e7037ac
2024-07-05 00:58:05 +08:00
hzhaoy
9ea80d83b8 tiny fix
Former-commit-id: 8f43ad988a4fd518a708fba53a173596ce2c59dd
2024-07-04 10:20:28 +08:00
hiyouga
4a590180d5 tiny fix
Former-commit-id: 935703b46d2871ce1014832da067dfe4a50c0610
2024-07-04 03:02:23 +08:00
hiyouga
a718f0eb51 fix data map for packing
Former-commit-id: ee6f8f926f084a195b2dbbd074e041e6c62c6ef4
2024-07-04 03:01:31 +08:00
hiyouga
a0df8be4e8 fix packing for eager/sdpa attn
Former-commit-id: 735a033ceb7f2da6da71d138ea091d8a665411a9
2024-07-04 01:52:43 +08:00
hoshi-hiyouga
9dcdaee09c Merge pull request #4224 from chuan298/main
Implement efficient packing without cross-contamination attention

Former-commit-id: ac382cc9fe4ec483658fd54f07f9a123788ce1b1
2024-07-04 01:18:54 +08:00
hiyouga
bd294e7cc3 update packing
Former-commit-id: f3d9c31efa0e64317bdd5b4ed6f78653cf3b5ba4
2024-07-04 01:10:55 +08:00
hiyouga
fe888a9073 update hparams
Former-commit-id: 1c4feac44192b1f540208837f5a530b0d3f5fb37
2024-07-03 23:18:58 +08:00
hiyouga
ca106d1f1b improve rlhf
Former-commit-id: e441780e3db256ca09a442ea9254e7ce16898a07
2024-07-02 22:23:08 +08:00
ancv
260f55ea47 move efficient_packing from data_args to model_args
Former-commit-id: 7b61659c707480bcf8c802c73e10d12ad5b9b965
2024-07-02 18:37:55 +07:00
hoshi-hiyouga
9174675ba9 Merge branch 'main' into main
Former-commit-id: 7be442f37d53a0c6324728fa1fa8e2c84d7f0fa5
2024-07-01 21:01:09 +08:00
hiyouga
a43f518389 fix #4402 #4617
Deprecate reserved_label_len arg


Former-commit-id: 4b6568984c0be4b31e7aa91b7c0d52b7f7b12b0b
2024-07-01 01:19:27 +08:00
hiyouga
0265a98942 fix #4556
Former-commit-id: 81faa9a985c14e83e38f42aedd228edb676b0695
2024-06-26 19:43:16 +08:00
hiyouga
f9d00ce051 tiny fix
Former-commit-id: bb57478366a70a0871af30ab31c890f471e27ff4
2024-06-25 01:15:19 +08:00
hoshi-hiyouga
1dcb792c82 Merge pull request #4417 from mMrBun/main
Add tool_format parameter to rewrite templates for different function call formats.

Former-commit-id: 8d1460cad5bff5e4626fdd675046021e0a3d1947
2024-06-24 23:17:55 +08:00
hoshi-hiyouga
d9529b749e Update template.py
Former-commit-id: d53517bff6f8734221d7df9982f3bdd4d2eb2cab
2024-06-24 23:12:59 +08:00