wassname/alignment-handbook
1
0
Fork 0
mirror of https://github.com/wassname/alignment-handbook.git synced 2026-06-27 15:58:47 +08:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • 6d128ea986 wip main wassname 2025-06-04 05:37:07 +00:00
  • a264efaa4c better formating wassname 2025-06-03 22:21:18 +00:00
  • 097e4e0b01 wip wassname 2025-06-02 22:31:52 +00:00
  • 880d4eda1e chat template fix wassname 2025-06-02 07:27:46 +00:00
  • 2819dd46d0 fmt wassname 2025-06-02 07:13:52 +00:00
  • fc7d4ed451 configs wassname 2025-06-02 06:20:04 +00:00
  • 8708597941 wip wassname 2025-06-02 05:51:13 +00:00
  • 205b881c80 Typo fixed (#206) Sergio Paniego Blanco 2025-04-30 16:19:36 +02:00
  • 5a7ba955d3 Pin TRL (#212) Bram Vanroy 2025-04-30 16:18:57 +02:00
  • ae3f44fc7d Add Smollm2 pipeline (#205) Loubna Ben Allal 2024-11-21 13:46:39 +01:00
  • e057d7f45b Change 'three' to 'four' in README (#202) Mikhail Poludin 2024-11-19 11:00:16 +01:00
  • 7d711cd80d Update model_utils.py for new API (#199) Kashif Rasul 2024-09-23 09:48:40 +02:00
  • 27f7dbf006 Fix BitsAndBytes JSON Serializable (#191) Chansung Park 2024-08-20 22:02:18 +09:00
  • a8dcde2cd3 Use ungated models for unit tests (#196) lewtun 2024-08-19 12:33:11 +02:00
  • 28bf902477 Use get_chat_template() instead of default_chat_template (#197) Alvaro Bartolome 2024-08-19 10:12:39 +02:00
  • 73dce0c35d Add Smollm (#194) Loubna Ben Allal 2024-08-19 08:47:20 +02:00
  • 01f29c1325 remove revision (#186) Kashif Rasul 2024-07-31 21:23:10 +02:00
  • 444e0f8414 Update README.md (#184) Kashif Rasul 2024-07-30 11:05:50 +02:00
  • 98563353d7 CITATION.cff and fix F401 warning (#183) Kashif Rasul 2024-07-30 10:56:17 +02:00
  • 95dc47218c update API to use latest TRL (#182) Kashif Rasul 2024-07-30 09:16:25 +02:00
  • 606d2e954f Add fsdp+qlora support (#160) Chansung Park 2024-05-08 22:08:13 +09:00
  • 84f8c92820 Update README.md (#152) Zizheng Yang 2024-04-25 16:36:22 +08:00
  • cf1975a7cb Add ORPO within README.md files (#154) Alvaro Bartolome 2024-04-25 10:35:45 +02:00
  • 70769f9e9b Add run_orpo.py (#143) Alvaro Bartolome 2024-04-11 16:02:20 +02:00
  • a83b1f617f Fix the logic that causes an issue with philschmid/gemma-tokenizer-chatml tokenizer (#146) kykim0 2024-04-09 08:02:21 -07:00
  • 8497caeaf1 fix trust_remote_code for tokenizer in model_utils.py (#140) Qingqing Cao 2024-03-27 11:31:45 -07:00
  • ba7e0e4fca Fix dataloading for cpt (#137) Bram Vanroy 2024-03-21 20:05:53 +01:00
  • c44cb1cd1d fix: Zephyr LoRA fine-tuning fixed (#139) Sergei Bogdanov 2024-03-21 19:28:31 +01:00
  • 595023faa4 Adding continued_pretraining task (#131) Bram Vanroy 2024-03-14 15:15:23 +01:00
  • a9b8a50a27 🌟 (#135) lewtun 2024-03-12 17:22:21 +01:00
  • ff618a4d13 🪁 (#129) lewtun 2024-03-01 17:29:42 +01:00
  • d17fd7cd3b Add auto_insert_empty_system_msg config flag (#123) Bram Vanroy 2024-02-28 20:05:44 +01:00
  • 87cc800498 Apply quantization during DPO QLoRA (#115) lewtun 2024-02-05 16:50:17 +01:00
  • d00e6f043e Update README.md (#113) Ikko Eltociear Ashimine 2024-02-02 17:20:20 +09:00
  • b4bd3a4984 Blog post url: "constitutional-ai" -> "constitutional_ai" (#112) Kosti 2024-02-01 17:21:53 +00:00
  • 995d50912b Update README.md (#111) lewtun 2024-02-01 17:02:43 +01:00
  • 8df2271324 Constitutional AI recipe (#108) Costa Huang 2024-02-01 07:02:19 -08:00
  • 5ad6db0c79 Fixes #96 by handling RepositoryNotFoundError (#97) Traun Leyden 2024-02-01 15:47:14 +01:00
  • ad3d43aeea Make peft bnb_4bit_compute_dtype consistent with torch_dtype (#107) Nathan Azrak 2024-01-29 19:59:15 +09:00
  • de7d8883cd Add check before inserting system message (#106) Nathan Azrak 2024-01-29 19:56:24 +09:00
  • cbcb3f60fb DPO/IPO/KTO ablations (#104) Edward Beeching 2024-01-18 14:55:00 +01:00
  • c74ed11171 Bump lower version of huggingface_hub (#95) lewtun 2024-01-11 23:09:48 +11:00
  • f0ffa0d7a6 Update Zephyr configs to account for UltraFeedback & TRL fixes (#88) lewtun 2024-01-10 17:42:24 +11:00
  • c69ae4b8a5 Check that default_chat_template is also None (#83) Nathan Azrak 2024-01-08 17:54:23 +11:00
  • 98fe28fb14 Clean deprecated max_samples arguments (#89) Kirill 2024-01-05 02:06:47 +04:00
  • e316174e1c Add warmup to config (#71) Evgenii Zheltonozhskii 2024-01-04 07:04:46 +02:00
  • 57508b5c2d Make SFT script consistent with DPO script (#86) NielsRogge 2024-01-04 05:55:58 +01:00
  • 8f6e5b666b Bump dev version v0.3.0.dev Lewis Tunstall 2024-01-04 01:39:11 +00:00
  • 61a11a5c7d Update docstring for data.py to reflect true behavior of shuffle parameter (#60) Scott Fleming 2023-12-06 01:44:17 -08:00
  • 1c06e4e5e1 Update doc CI (#64) lewtun 2023-12-05 12:31:30 +01:00
  • 3f368a0748 Add check that parameters are not intended to be offloaded (#51) Nathan Azrak 2023-12-04 19:10:41 +11:00
  • 15279e7157 Allow loading datasets from disk using load_from_disk method. (#53) Dragan Milchevski 2023-12-01 11:05:35 +01:00
  • 80e952ec47 Allow running DPO from a local model (#49) Dragan Milchevski 2023-11-27 11:31:09 +01:00
  • f025057ce4 Missing config params on SFT (#31) Thomas Capelle 2023-11-21 12:00:09 +01:00
  • c9d9035f95 Fix apply_chat_template function for dpo and unknown task (#30) Alvaro Bartolome 2023-11-21 11:47:21 +01:00
  • 7d6fe765ec Update README.md (#35) Girraj Jangid 2023-11-20 13:22:16 +05:30
  • 0e09b0c6ec Fix note syntax highlighting in README.md (#20) Alvaro Bartolome 2023-11-15 08:45:27 +01:00
  • a1afb2bbd4 Fix image alignment (#19) lewtun 2023-11-12 15:47:10 +01:00
  • 4c6226bc42 Add moar explanations (#18) lewtun 2023-11-12 15:43:39 +01:00
  • 4b0c1fe170 fix typos (#17) Kashif Rasul 2023-11-12 13:44:50 +01:00
  • 43f52224db Merge pull request #14 from sebastianschramm/ses/fix_typos_zephyr_recipe lewtun 2023-11-10 17:06:35 +01:00
  • d48a4a477b Resolves #13 fix typo in zephyr recipe readme Sebastian Schramm 2023-11-10 16:57:39 +01:00
  • e4f98e7d8f Merge pull request #11 from huggingface/zephyr-recipe lewtun 2023-11-10 15:54:13 +01:00
  • 363e29ff95 Apply suggestions from code review lewtun 2023-11-10 15:49:00 +01:00
  • f5e70fbf9e Add licenses Lewis Tunstall 2023-11-10 14:47:54 +00:00
  • 5a630a1989 Add QLoRA command Lewis Tunstall 2023-11-10 13:57:52 +00:00
  • e2e8ab945d Refactor imports Lewis Tunstall 2023-11-10 13:38:45 +00:00
  • 7f1a14e0d4 adds auto adapter merge to dpo script edbeeching 2023-11-10 14:15:44 +01:00
  • 54185783e0 Remove QLoRa for now Lewis Tunstall 2023-11-10 11:20:39 +00:00
  • edf67d1d93 Tweaks Lewis Tunstall 2023-11-10 11:15:45 +00:00
  • 551f901f95 Fix dep Lewis Tunstall 2023-11-10 11:02:44 +00:00
  • a0b8d49424 Rename recipe Lewis Tunstall 2023-11-10 10:49:13 +00:00
  • 64f1834e01 Add config tests Lewis Tunstall 2023-11-10 10:00:05 +00:00
  • 8699f47bf3 Add jinja2 to req deps Lewis Tunstall 2023-11-10 09:45:22 +00:00
  • b1b0c1c8c0 Update setup.py lewtun 2023-11-10 10:44:06 +01:00
  • 2ed5a45d25 Add model utils tests Lewis Tunstall 2023-11-10 09:42:15 +00:00
  • 0af8011993 Bump deps Lewis Tunstall 2023-11-10 08:41:17 +00:00
  • 610a1a2de4 Add unit tests for data mixer Lewis Tunstall 2023-11-10 08:37:53 +00:00
  • 0f0b61c096 ups lora bs x grad_acc to 64 edbeeching 2023-11-10 09:30:54 +01:00
  • 13141a4b0b adds updated model paths, adds eval to sft scripts edbeeching 2023-11-10 09:26:39 +01:00
  • 4b0769d137 Fix links Lewis Tunstall 2023-11-09 14:42:57 +00:00
  • 89f58a043c Add project structure Lewis Tunstall 2023-11-09 14:40:23 +00:00
  • 44b324487d Bump bs Lewis Tunstall 2023-11-09 14:20:43 +00:00
  • 756bb76d22 Fix Slurm opts Lewis Tunstall 2023-11-09 14:09:52 +00:00
  • 33a0ce3afd Add more doc Lewis Tunstall 2023-11-09 13:39:03 +00:00
  • 3a5430222e removes need for yq dep edbeeching 2023-11-09 13:04:34 +01:00
  • 49da3ef739 adds configs and instructions for lora training edbeeching 2023-11-09 10:56:25 +01:00
  • 2de17f5ba1 Add doc Lewis Tunstall 2023-11-09 07:32:24 +00:00
  • e2c19a0252 Tweak Lewis Tunstall 2023-11-08 23:09:16 +00:00
  • ee10c4efd9 Make DPO work! Lewis Tunstall 2023-11-08 22:58:34 +00:00
  • e54e095978 Make it work for realz Lewis Tunstall 2023-11-08 22:20:17 +00:00
  • d2900adc83 Make it work! Lewis Tunstall 2023-11-08 16:31:57 +00:00
  • 967eab4cfb Add skeleton Lewis Tunstall 2023-11-08 13:21:57 +00:00
  • b9d9aa0a29 Fix style Lewis Tunstall 2023-10-30 10:00:43 +01:00
  • 3d8570af1e Update README.md lewtun 2023-10-26 23:21:53 +02:00
  • da5dfbe9b6 Fix tests Lewis Tunstall 2023-10-26 17:50:31 +00:00
  • a28b4cfc6e Bump dev version Lewis Tunstall 2023-10-26 10:17:28 +00:00
  • 1ca8add5fa Add release details Lewis Tunstall 2023-10-26 10:10:57 +00:00
  • 87033c09b3 Add tests folder Lewis Tunstall 2023-10-26 09:54:20 +00:00
  • 1bde6a7931 Add doc builder Lewis Tunstall 2023-10-26 09:41:04 +00:00