hiyouga
|
88745c9bb5
|
fix #3873
Former-commit-id: 91611d68c4
|
2024-06-04 00:21:50 +08:00 |
|
hiyouga
|
8ecf606230
|
fix #3992
Former-commit-id: a18acf2abe
|
2024-06-04 00:17:36 +08:00 |
|
hiyouga
|
b12d4beb8a
|
fix abort in webui DDP mode
Former-commit-id: 2187518762
|
2024-06-04 00:10:24 +08:00 |
|
hoshi-hiyouga
|
326f180397
|
Merge pull request #3987 from injet-zhou/main
Fix cann't interrupt training when using multi GPUs in webui
Former-commit-id: ae18e1e251
|
2024-06-04 00:04:07 +08:00 |
|
hiyouga
|
e2920aa925
|
fix #4043
Former-commit-id: 79784ebeb6
|
2024-06-03 23:30:37 +08:00 |
|
hiyouga
|
6f7b6ae0c3
|
remove gc warnings in DPO&KTO
Former-commit-id: f9a206509e
|
2024-06-03 22:53:54 +08:00 |
|
hoshi-hiyouga
|
b2c224de69
|
Merge pull request #4045 from enji-zhou/feature/add_kto
fix KTO Trainer Sampler
Former-commit-id: 30a538e2db
|
2024-06-03 22:09:25 +08:00 |
|
hoshi-hiyouga
|
5d96cf146e
|
Update trainer.py
Former-commit-id: 24499f40dc
|
2024-06-03 22:08:38 +08:00 |
|
enji.zhou
|
e58aca0602
|
fix KTO Trainer Sampler
Former-commit-id: 34a2c5087a
|
2024-06-03 21:32:38 +08:00 |
|
hoshi-hiyouga
|
f6f1c4eacb
|
Merge pull request #4006 from Uminosachi/scheduler-kwargs
Set scheduler_specific_kwargs to get_scheduler
Former-commit-id: 0f01500b68
|
2024-06-03 19:27:53 +08:00 |
|
hiyouga
|
a187068e7c
|
update placeholder in issue template
Former-commit-id: 88681d3357
|
2024-06-03 19:24:10 +08:00 |
|
hoshi-hiyouga
|
cdfd2ad4b1
|
Merge pull request #4011 from statelesshz/issue-template
Update bug-report.yml
Former-commit-id: d359dd2de4
|
2024-06-03 19:20:43 +08:00 |
|
hiyouga
|
e4ce59243b
|
fix #4005 #4013
Former-commit-id: eed33862bc
|
2024-06-03 19:12:29 +08:00 |
|
hoshi-hiyouga
|
eaab09fccb
|
Merge pull request #4007 from xu-song/patch-3
Update model_args.py
Former-commit-id: 1539c72b94
|
2024-06-03 18:54:37 +08:00 |
|
hiyouga
|
d0ceb1b091
|
fix #4022
Former-commit-id: 24e1c0e2ee
|
2024-06-03 18:38:36 +08:00 |
|
hiyouga
|
af7748139a
|
bump versions
transformers 4.37.2->4.41.2
datasets 2.14.3->2.16.0
accelerate 0.27.2->0.30.1
peft 0.10.0->0.11.1
trl 0.8.1->0.8.6
Former-commit-id: 876bc92865
|
2024-06-03 18:29:38 +08:00 |
|
hiyouga
|
64d24842fe
|
fix data loader hint
Former-commit-id: 49b1e88e3d
|
2024-06-03 18:28:27 +08:00 |
|
ylfeng
|
62d55b71a3
|
remove empty line
Former-commit-id: b47e317447
|
2024-05-31 21:43:08 +08:00 |
|
ylfeng
|
0feb2ad35c
|
fix eos
Former-commit-id: 84aee57901
|
2024-05-31 21:40:41 +08:00 |
|
ylfeng
|
8350e508d3
|
supervised packing with greedy knapsack algorithm
Former-commit-id: f9db439cb7
|
2024-05-31 15:33:54 +08:00 |
|
Xu Song
|
abe33220bf
|
Update model_args.py
Former-commit-id: dade2f083d
|
2024-05-31 14:35:48 +08:00 |
|
statelesshz
|
6a6f07053d
|
Update bug-report.yml
Former-commit-id: f78e21f341
|
2024-05-31 13:18:18 +08:00 |
|
Uminosachi
|
0de4e1e9e2
|
Set scheduler_specific_kwargs to get_scheduler
Former-commit-id: 14e97dc119
|
2024-05-31 13:45:39 +09:00 |
|
hiyouga
|
72ebcb9a04
|
update readme
Former-commit-id: c4f50865ad
|
2024-05-30 16:40:17 +08:00 |
|
faddddeout
|
64976e426c
|
fix cann't interrupt training when using multi GPUs in webui
Former-commit-id: b13d03946e
|
2024-05-30 08:39:21 +00:00 |
|
hoshi-hiyouga
|
e24276cab6
|
Update wechat.jpg
Former-commit-id: 2f38c1f5fd
|
2024-05-30 12:48:47 +08:00 |
|
hiyouga
|
107e39f2de
|
fix #3837
Former-commit-id: 3404e8f302
|
2024-05-30 00:52:26 +08:00 |
|
hoshi-hiyouga
|
9b6bdf9449
|
Merge pull request #3829 from seanzhang-zhichen/add_dataset_sample_num
Add dataset sample num
Former-commit-id: 483eb47e5d
|
2024-05-30 00:25:45 +08:00 |
|
hoshi-hiyouga
|
7b83c550ab
|
Update loader.py
Former-commit-id: ca5dd7c6c1
|
2024-05-30 00:20:20 +08:00 |
|
hoshi-hiyouga
|
9fc713da89
|
Update loader.py
Former-commit-id: f9a88b89ca
|
2024-05-30 00:17:21 +08:00 |
|
hoshi-hiyouga
|
c0f11a280e
|
Update loader.py
Former-commit-id: b55fb611c5
|
2024-05-30 00:12:12 +08:00 |
|
hoshi-hiyouga
|
69a51cacb1
|
Update parser.py
Former-commit-id: 51dd454337
|
2024-05-30 00:05:20 +08:00 |
|
hoshi-hiyouga
|
21e7979837
|
Update README_zh.md
Former-commit-id: c8ae7e0e65
|
2024-05-30 00:04:47 +08:00 |
|
hoshi-hiyouga
|
eb7ee82f16
|
Update README.md
Former-commit-id: 3761d7d5dd
|
2024-05-30 00:04:26 +08:00 |
|
hiyouga
|
820404946e
|
better llamaboard
* easily resume from checkpoint
* support full and freeze checkpoints
* faster ui
Former-commit-id: 8070871732
|
2024-05-29 23:55:38 +08:00 |
|
hiyouga
|
19a3262387
|
fix cohere system
Former-commit-id: d0aa36b8ad
|
2024-05-29 20:58:23 +08:00 |
|
hiyouga
|
c05cb3769f
|
fix #3965
Former-commit-id: 0930f58699
|
2024-05-29 20:55:51 +08:00 |
|
hiyouga
|
a71a6a05c3
|
update readme
Former-commit-id: 89ca832740
|
2024-05-29 18:39:11 +08:00 |
|
hoshi-hiyouga
|
2e7dae0f97
|
Merge pull request #3930 from MengqingCao/npu
Add Ascend npu doc and dependency
Former-commit-id: 880b4a9acf
|
2024-05-29 18:33:38 +08:00 |
|
MengqingCao
|
2192616770
|
update torch-npu version
Former-commit-id: b19d33b9d2
|
2024-05-29 10:05:11 +00:00 |
|
MengqingCao
|
29fe1cd688
|
update cann kernels url
Former-commit-id: e14f5b37e4
|
2024-05-29 09:53:31 +00:00 |
|
hoshi-hiyouga
|
10ab2861d5
|
Merge pull request #3958 from hzhaoy/add_telechat_12b_support
add TeleChat-12B/TeleChat-12B-v2 models
Former-commit-id: bcddd5b066
|
2024-05-29 17:20:53 +08:00 |
|
hzhaoy
|
ce1be3da4b
|
add TeleChat-12B/TeleChat-12B-v2 models
Former-commit-id: 0dd632fe9e
|
2024-05-29 15:00:37 +08:00 |
|
codemayq
|
2540ce58c0
|
update wechat_npu
Former-commit-id: 4728fe62d1
|
2024-05-29 11:13:59 +08:00 |
|
hiyouga
|
05277ee864
|
fix hf chat engine
Former-commit-id: 97346c1d3d
|
2024-05-29 01:20:07 +08:00 |
|
hiyouga
|
13e7b64641
|
add ds config to webui
Former-commit-id: e4b420c146
|
2024-05-29 01:13:17 +08:00 |
|
hiyouga
|
468d0e7ed1
|
10x generate in ppo w/ zero3
https://github.com/huggingface/trl/pull/1483
Former-commit-id: 65cd8bdbdb
|
2024-05-29 00:23:23 +08:00 |
|
hiyouga
|
bfac965f9c
|
update dpo, kto trainer
Former-commit-id: 7c8e01bb74
|
2024-05-29 00:14:29 +08:00 |
|
hiyouga
|
14f6cc2b7c
|
clean kto trainer
Former-commit-id: 900e1ea622
|
2024-05-28 21:43:26 +08:00 |
|
hiyouga
|
87e71df597
|
bump vllm version to 0.4.1
Former-commit-id: 1e80a3a638
|
2024-05-28 21:27:27 +08:00 |
|