Commit Graph

  • 63da6294dd Merge branch 'hiyouga:main' into main Johann-Peter Hartmann 2024-02-04 12:51:25 +00:00
  • db0ab4d601 fix reserved label len hiyouga 2024-02-04 17:54:26 +08:00
  • 19d33ede13 fix #2420 hiyouga 2024-02-04 15:51:47 +08:00
  • b988ce0a0c fix #2189 hiyouga 2024-02-04 00:47:37 +08:00
  • 38e63bfd28 bump up transformers version hiyouga 2024-02-04 00:01:16 +08:00
  • 51df865734 fix #2397 hiyouga 2024-02-03 23:45:31 +08:00
  • 6545c02790 add hint for freeze #2412 hiyouga 2024-02-03 23:38:56 +08:00
  • 4ecadc3512 fix #2376 hiyouga 2024-02-03 23:14:31 +08:00
  • 901faa16cc support minicpm #2404 hiyouga 2024-02-03 22:36:46 +08:00
  • b0ffde6e98 add simple german chatml template chatml_de Johann-Peter Hartmann 2024-02-03 09:01:15 +01:00
  • 870182c3a9 remove comma Johann-Peter Hartmann 2024-02-03 08:48:39 +01:00
  • 5318cc650e Merge branch 'hiyouga:main' into main Johann-Peter Hartmann 2024-02-03 08:43:12 +01:00
  • 2eaaa9522f Merge pull request #2411 from lxsyz/main hoshi-hiyouga 2024-02-02 17:38:16 +08:00
  • 3399c0d645 fix eos_token_id=0 bug Fallen Angel 2024-02-02 17:34:48 +08:00
  • a5d559a89e Merge branch 'hiyouga:main' into main Johann-Peter Hartmann 2024-02-01 07:13:31 +01:00
  • ec70eb5d8d Update wechat.jpg hiyouga 2024-02-01 11:01:12 +08:00
  • 4e27950acb Merge branch 'hiyouga:main' into main Johann-Peter Hartmann 2024-01-31 14:05:52 +01:00
  • 39bd5bd524 fix #2388 hiyouga 2024-01-31 17:23:56 +08:00
  • 521ad76552 fix autoset attn impl, update data readme hiyouga 2024-01-31 11:58:07 +08:00
  • d9a8301ed4 Add support for german datasets Johann-Peter Hartmann 2024-01-30 10:18:01 +01:00
  • 60733275f8 Update wechat.jpg hiyouga 2024-01-25 13:02:49 +08:00
  • 2bc30763e9 fix #2320 hiyouga 2024-01-24 16:19:18 +08:00
  • 51ad35b3c7 Merge pull request #2319 from ftgreat/main hoshi-hiyouga 2024-01-24 15:32:26 +08:00
  • c284665425 Add patch_mixtral_replace_moe_impl for full training Mitral using DeepSpeed Zero3. ldwang 2024-01-24 15:25:31 +08:00
  • 18923b1402 Add patch_mixtral_replace_moe_impl for full training Mitral using DeepSpeed Zero3. ldwang 2024-01-24 14:43:16 +08:00
  • dbaaa4546e Update dataset_info.json hiyouga 2024-01-23 00:10:32 +08:00
  • e4ba1deedf add hint hiyouga 2024-01-22 23:32:01 +08:00
  • 8e4b3a959a Merge pull request #2283 from A-Cepheus/main hoshi-hiyouga 2024-01-22 23:28:45 +08:00
  • bdc9eff635 Update patcher.py hoshi-hiyouga 2024-01-22 23:27:39 +08:00
  • bb5a3b85a6 Update tests.yml hoshi-hiyouga 2024-01-22 23:22:15 +08:00
  • 087476da99 Create tests.yml hoshi-hiyouga 2024-01-22 23:13:04 +08:00
  • b2fb0eca56 fix #2282 and update tool prompt hiyouga 2024-01-22 22:27:30 +08:00
  • 6fc2d5cc03 add orion models hiyouga 2024-01-22 21:26:53 +08:00
  • b06a31e76a 🐞 fix: typo A-Cepheus 2024-01-22 16:04:39 +08:00
  • 319a72b48d 🐞 fix: typo, move MoE fix to patcher A-Cepheus 2024-01-22 16:01:58 +08:00
  • e1d5c98519 fix: ZeRO3 does not work with MoE models A-Cepheus 2024-01-22 15:21:14 +08:00
  • 486cc8d360 add array param format hiyouga 2024-01-21 22:17:48 +08:00
  • 37ec5deaec update tool test hiyouga 2024-01-21 19:41:46 +08:00
  • f51d48ea08 fix api hiyouga 2024-01-21 19:15:27 +08:00
  • e0a717aa3a fix #2268 hiyouga 2024-01-21 14:11:38 +08:00
  • 3bc301be02 tiny fix hiyouga 2024-01-21 13:26:12 +08:00
  • ea6db72631 Merge pull request #2266 from yhyu13/fix_export_model_dtype hoshi-hiyouga 2024-01-21 12:40:39 +08:00
  • 2533be5e2a Merge branch 'main' into fix_export_model_dtype hoshi-hiyouga 2024-01-21 12:40:24 +08:00
  • 662b9a9dcf Update tuner.py hoshi-hiyouga 2024-01-21 12:39:38 +08:00
  • 11f612deab Merge pull request #2262 from fenglui/main hoshi-hiyouga 2024-01-21 12:34:37 +08:00
  • 3c1e256992 format hiyouga 2024-01-21 12:34:17 +08:00
  • 3ba75dc071 Merge pull request #2264 from seoeaa/main hoshi-hiyouga 2024-01-21 12:25:24 +08:00
  • 9cdbd3bfc8 Remove manully set use_cache; torch_dtype is not str, save model as bfloat16 used to fail; yhyu13 2024-01-21 11:12:15 +08:00
  • ace7fd4e0a add russian lang Aleksandr 2024-01-21 04:28:14 +03:00
  • f7c5b2d105 fix torch_dtype check of export_model fenglui 2024-01-21 05:01:53 +08:00
  • a0d59aa4ec release v0.5.0 (real) hiyouga 2024-01-21 01:54:49 +08:00
  • 3e982cc714 finish agent hiyouga 2024-01-21 01:47:33 +08:00
  • 55f707196e fix api hiyouga 2024-01-21 00:03:09 +08:00
  • a9c18255aa fix internlm2 template hiyouga 2024-01-20 23:33:50 +08:00
  • c550987a72 fix cli_demo hiyouga 2024-01-20 23:27:10 +08:00
  • cf818a2598 fix #2260 hiyouga 2024-01-20 23:22:09 +08:00
  • 71cfdc2658 release v0.5.0 hiyouga 2024-01-20 20:21:39 +08:00
  • 638234ceee format style hiyouga 2024-01-20 20:15:56 +08:00
  • f6d6e00337 fix tests hiyouga 2024-01-20 19:58:04 +08:00
  • 38af076a75 support longlora for main branch hiyouga 2024-01-20 19:25:22 +08:00
  • bb92cdd0db Merge pull request #2201 from liu-zichen/token_embed_resize hoshi-hiyouga 2024-01-20 17:45:38 +08:00
  • 8cbe4e9609 add upcast_lmhead option hiyouga 2024-01-19 23:54:25 +08:00
  • 0ff9a1fb4f set use_reentrant=False hiyouga 2024-01-19 23:29:54 +08:00
  • 12043aab9c fix #2249 hiyouga 2024-01-19 21:44:32 +08:00
  • b6ec112beb add bf16 lora option hiyouga 2024-01-19 16:29:03 +08:00
  • 35aef8b287 fix function formatter hiyouga 2024-01-18 16:01:07 +08:00
  • ddd48ce8ab Update tuner.py hiyouga 2024-01-18 15:06:02 +08:00
  • b5ef993e34 update wechat hiyouga 2024-01-18 14:54:38 +08:00
  • a73a979afd fix templates hiyouga 2024-01-18 14:49:52 +08:00
  • 5edf7cce0e fix rm dataset hiyouga 2024-01-18 14:45:37 +08:00
  • 5ff10fac4f fix pretrain data loader hiyouga 2024-01-18 14:42:52 +08:00
  • 9986cc6dd1 Merge pull request #2226 from hiyouga/dev hoshi-hiyouga 2024-01-18 14:31:28 +08:00
  • 5608a0da8e update readme hiyouga 2024-01-18 14:30:48 +08:00
  • 2abfe5fbc2 add tool hint hiyouga 2024-01-18 13:19:09 +08:00
  • 487dee066f fix dataset hiyouga 2024-01-18 12:59:30 +08:00
  • f1067d2b58 enable cutoff len hiyouga 2024-01-18 12:25:42 +08:00
  • 83dbfce8c3 add tool test hiyouga 2024-01-18 10:26:26 +08:00
  • d9f1cae351 support function calling hiyouga 2024-01-18 09:54:23 +08:00
  • 28135d787d Update llamafy_internlm2.py hiyouga 2024-01-18 01:12:31 +08:00
  • 484becae1b Update llamafy_internlm2.py hiyouga 2024-01-18 01:00:16 +08:00
  • c84a387c2c Update llamafy_internlm2.py hiyouga 2024-01-18 00:49:31 +08:00
  • f99140d5e8 fix llamafy scripts hiyouga 2024-01-18 00:37:37 +08:00
  • 7ff4c874d2 fix llamafy_internlm2 hiyouga 2024-01-18 00:26:14 +08:00
  • f1d7ca77b1 add llamafy_internlm2 hiyouga 2024-01-18 00:17:41 +08:00
  • 42859f0734 support export push_to_hub #2183 hiyouga 2024-01-16 23:59:42 +08:00
  • a83fb6d3ff fix #2195 hiyouga 2024-01-16 23:53:50 +08:00
  • a5f6a7f4fb support resize embed for zero3 liuzc 2024-01-16 15:16:20 +08:00
  • 5a207bb723 tiny fix hiyouga 2024-01-15 23:34:23 +08:00
  • 3aa8901994 Merge pull request #2194 from junuMoon/patch-1 hoshi-hiyouga 2024-01-15 20:21:28 +08:00
  • 7a320de097 fix: typo on README.md Junu Moon(Fran) 2024-01-15 19:50:35 +09:00
  • bf73224f33 support solar 10.7B #1907 hiyouga 2024-01-14 00:30:30 +08:00
  • 3c8e72f585 Update README_zh.md hiyouga 2024-01-14 00:17:28 +08:00
  • ca3933dc52 support deepseek moe hiyouga 2024-01-14 00:14:49 +08:00
  • d1a73fe26c fix phi modules hiyouga 2024-01-13 23:12:47 +08:00
  • 9aa1a2fc17 fix #2147 hiyouga 2024-01-12 03:30:56 +08:00
  • 4b2d11ec28 fix #2164 hiyouga 2024-01-12 00:27:57 +08:00
  • 7bf6612f4a Merge pull request #2163 from JessyTsu1/main hoshi-hiyouga 2024-01-11 23:33:29 +08:00
  • 8c5e4a8896 Update README.md JessyTsu1 2024-01-11 23:18:29 +08:00
  • cdeca0cabc Update README_zh.md JessyTsu1 2024-01-11 23:17:48 +08:00
  • d72aff5ae6 Update README.md JessyTsu1 2024-01-11 23:17:00 +08:00