Commit Graph

  • e2cfd34da0 update torch-npu version MengqingCao 2024-05-29 10:05:11 +00:00
  • 668dea9706 update cann kernels url MengqingCao 2024-05-29 09:53:31 +00:00
  • 084be442f2 Merge pull request #3958 from hzhaoy/add_telechat_12b_support hoshi-hiyouga 2024-05-29 17:20:53 +08:00
  • 29cb4a1327 add TeleChat-12B/TeleChat-12B-v2 models hzhaoy 2024-05-29 15:00:37 +08:00
  • 81a61134b8 fix hf chat engine hiyouga 2024-05-29 01:20:07 +08:00
  • cb1a49aa02 add ds config to webui hiyouga 2024-05-29 01:13:17 +08:00
  • 351b4efc6c 10x generate in ppo w/ zero3 hiyouga 2024-05-29 00:23:23 +08:00
  • 9b551309de update dpo, kto trainer hiyouga 2024-05-29 00:14:29 +08:00
  • 9fed4a2ef4 clean kto trainer hiyouga 2024-05-28 21:43:26 +08:00
  • bceac4f554 bump vllm version to 0.4.1 hiyouga 2024-05-28 21:27:27 +08:00
  • ae3a88d3a7 update readme hiyouga 2024-05-28 19:35:52 +08:00
  • 9138a7a5ba support DDP in webui hiyouga 2024-05-28 19:24:22 +08:00
  • 9912b43fcc update readme hiyouga 2024-05-28 16:41:34 +08:00
  • 5ac37555a4 update readme hiyouga 2024-05-28 16:19:56 +08:00
  • 34bdc730a6 fix #3931 hiyouga 2024-05-28 13:44:22 +08:00
  • e45a9d70fc add Ascend npu doc and dependency MengqingCao 2024-05-28 01:33:54 +00:00
  • 232b36059c Merge pull request #3925 from Yimi81/feat-fix-yi-template hoshi-hiyouga 2024-05-27 22:59:32 +08:00
  • d9fbd675d5 fix yi template Yimi81 2024-05-27 13:11:25 +00:00
  • 0206e7b9de tiny fix hiyouga 2024-05-27 20:54:26 +08:00
  • a886544d3d Merge pull request #3921 from gusye1234/main hoshi-hiyouga 2024-05-27 20:52:37 +08:00
  • 8c9b929bb0 Update template.py hoshi-hiyouga 2024-05-27 20:51:56 +08:00
  • 1bb1ae834e Update template.py hoshi-hiyouga 2024-05-27 20:51:26 +08:00
  • 0d9e364a90 add openchat-3.6-8B support Jianbai Ye 2024-05-27 20:42:08 +08:00
  • 3b28c003dd fix full/freeze tuning for mllm hiyouga 2024-05-27 20:37:57 +08:00
  • 48ff9fb150 Merge pull request #3835 from BUAADreamer/main hoshi-hiyouga 2024-05-27 20:23:45 +08:00
  • c43bc74fe6 support Aya23 hiyouga 2024-05-27 20:23:24 +08:00
  • eaf9cc2195 Merge branch 'hiyouga:main' into main BUAADreamer 2024-05-27 20:10:58 +08:00
  • 4bd276f58f add llava 1k datasets hiyouga 2024-05-27 19:57:33 +08:00
  • f8cf0d5e5d update dpo examples hiyouga 2024-05-27 19:56:04 +08:00
  • 79bc60db33 Merge branch 'hiyouga:main' into main BUAADreamer 2024-05-27 19:00:48 +08:00
  • dc7c54067e add only tune lm and mm_proj BUAADreamer 2024-05-27 19:00:15 +08:00
  • 932f0d5c20 add regex of only tune lm and mm_proj BUAADreamer 2024-05-27 18:59:00 +08:00
  • 9670f5e41a add phi-3 7b/14b, mistral v0.3 models hiyouga 2024-05-27 18:20:16 +08:00
  • 97a23e1cbe update readme hiyouga 2024-05-27 18:14:02 +08:00
  • 11fcd055ec Merge branch 'hiyouga:main' into main BUAADreamer 2024-05-27 11:54:01 +08:00
  • b0d9966663 support SimPO #3900 hiyouga 2024-05-26 23:46:33 +08:00
  • 5c51ab7e1f Merge branch 'hiyouga:main' into main BUAADreamer 2024-05-25 14:18:49 +08:00
  • 26f293d587 fix #3853 hiyouga 2024-05-24 23:29:45 +08:00
  • a3b52fd380 Merge branch 'main' into add_dataset_sample_num seanzhang-zhichen 2024-05-24 15:57:47 +08:00
  • 27d8706d6d Merge branch 'hiyouga:main' into main BUAADreamer 2024-05-24 09:50:00 +08:00
  • bf59383783 refactor data preprocessing, fix mllm rlhf hiyouga 2024-05-24 04:08:25 +08:00
  • 1078611259 Merge pull request #3876 from dongdongqiang2018/main hoshi-hiyouga 2024-05-24 01:54:30 +08:00
  • e6fc0ac8fe fix paligemma sft hiyouga 2024-05-24 00:23:40 +08:00
  • 554ca3d8dc fix oom issues in export hiyouga 2024-05-23 23:32:45 +08:00
  • 86dfdf956d adapted to 910B image donggang 2024-05-23 09:48:22 +00:00
  • c0e4475485 Merge branch 'hiyouga:main' into main BUAADreamer 2024-05-21 22:18:20 +08:00
  • 2b65f8bd5c fix paligemma sft hiyouga 2024-05-21 20:03:09 +08:00
  • 09e78272c2 Update README_zh.md hiyouga 2024-05-21 18:30:59 +08:00
  • cccce564bd update wechat hiyouga 2024-05-21 18:22:32 +08:00
  • 4adec327de fix #3847 hiyouga 2024-05-21 17:53:06 +08:00
  • 1f093334d1 support pretraining of llava BUAADreamer 2024-05-21 08:57:14 +08:00
  • e0e8507108 support paligemma hiyouga 2024-05-21 00:01:22 +08:00
  • f5962f8128 fix paligemma data preprocess hiyouga 2024-05-20 23:51:32 +08:00
  • b31d808655 fix paligemma inference hiyouga 2024-05-20 23:36:43 +08:00
  • 247cda4b68 fix #3818 hiyouga 2024-05-20 21:43:19 +08:00
  • e30975e9a2 add kto to webui hiyouga 2024-05-20 21:20:25 +08:00
  • de9f1583c2 fix conflict zhangzc 2024-05-20 17:10:01 +08:00
  • ab48653e63 fix chat engines hiyouga 2024-05-20 00:36:43 +08:00
  • 6d7a1e3f8f Merge pull request #3812 from ycjcl868/feat/chat-support-system-prompt hoshi-hiyouga 2024-05-20 00:31:32 +08:00
  • e093dad7cb Update vllm_engine.py hoshi-hiyouga 2024-05-20 00:31:04 +08:00
  • b103a121f0 Update hf_engine.py hoshi-hiyouga 2024-05-20 00:30:45 +08:00
  • 3578abc7a4 Update generating_args.py hoshi-hiyouga 2024-05-20 00:29:31 +08:00
  • 17d398f419 Update chat_model.py hoshi-hiyouga 2024-05-20 00:29:12 +08:00
  • 3453a8eebb fix jinja template hiyouga 2024-05-19 23:38:30 +08:00
  • 77a089c35c feat: cli chat support system_message ycjcl868 2024-05-19 23:17:46 +08:00
  • 516d83c946 fix zero2 high ram usage hiyouga 2024-05-19 21:53:54 +08:00
  • fd02c9f973 fix hf gen args hiyouga 2024-05-19 19:39:32 +08:00
  • 351e80a656 fix envs hiyouga 2024-05-19 18:27:18 +08:00
  • 4f04e2ed93 fix #3807 hiyouga 2024-05-19 17:07:57 +08:00
  • a810d1b98e update readme hiyouga 2024-05-18 23:09:03 +08:00
  • fbe963a96a safe output path in webui hiyouga 2024-05-18 22:42:28 +08:00
  • d13b8bee8a fix jetmoe z3 block hiyouga 2024-05-18 22:28:45 +08:00
  • 0aa072a155 improve data process logger hiyouga 2024-05-18 22:02:42 +08:00
  • 57dde7c3bc update data readme hiyouga 2024-05-18 21:37:38 +08:00
  • 6b9003f781 update data readme hiyouga 2024-05-18 21:15:20 +08:00
  • 9c1c59e481 fix #3803 hiyouga 2024-05-18 16:13:14 +08:00
  • 31daec2749 Merge pull request #3799 from hiyouga/dev hoshi-hiyouga 2024-05-18 03:49:13 +08:00
  • 2bff90719b improve KTO impl., replace datasets hiyouga 2024-05-18 03:44:56 +08:00
  • e4570e28a8 Merge pull request #3785 from enji-zhou/feature/add_kto hoshi-hiyouga 2024-05-18 03:07:18 +08:00
  • d84a730daa Merge pull request #3794 from jue-jue-zi/main hoshi-hiyouga 2024-05-17 16:17:30 +08:00
  • 0fd1a05cec Update model_args.py hoshi-hiyouga 2024-05-17 16:16:41 +08:00
  • 6373d307ec feat: pass the max_lora_rank parameter to vLLM backend juejuezi 2024-05-17 16:07:39 +08:00
  • a32c3a50fc add deepseek v2 lite model hiyouga 2024-05-17 13:25:36 +08:00
  • 66b5634ebf add kto enji.zhou 2024-05-17 13:09:17 +08:00
  • 92b3697e2c update badam example #3764 hiyouga 2024-05-17 02:21:10 +08:00
  • 969e605c7e better dtype handle in loading hiyouga 2024-05-17 02:14:56 +08:00
  • a3320f26cf update examples hiyouga 2024-05-17 01:02:00 +08:00
  • 45329d9e3c enable inbrowser in webui hiyouga 2024-05-17 00:08:56 +08:00
  • 6481321470 add falcon 11b hiyouga 2024-05-17 00:08:33 +08:00
  • efcf5e050d fix examples #3769 hiyouga 2024-05-16 19:12:09 +08:00
  • dfa686b617 rename package hiyouga 2024-05-16 18:39:08 +08:00
  • fe638cf11f set dev version hiyouga 2024-05-16 02:17:31 +08:00
  • b2949b88e9 release v0.7.1 v0.7.1 hiyouga 2024-05-16 00:57:16 +08:00
  • 538c79fd8f fix #3694 hiyouga 2024-05-16 00:35:28 +08:00
  • 437cc20be6 fix #3606 hiyouga 2024-05-15 23:05:02 +08:00
  • 2ac972d6e7 add Yi-VL-34B model hiyouga 2024-05-15 22:58:19 +08:00
  • 4d7f0fbb7a add yi-vl 6b model hiyouga 2024-05-15 20:02:41 +08:00
  • 40e3d3fbdd fix yi vl vllm infer hiyouga 2024-05-15 19:25:48 +08:00
  • 096677b989 add NPU docker images hiyouga 2024-05-15 19:20:11 +08:00
  • 7940b968ae Merge pull request #3748 from BUAADreamer/main hoshi-hiyouga 2024-05-15 16:40:54 +08:00