Commit Graph

  • 2ee7f5d5ee remove the clone method as it's twice the size? main deep1 2023-04-22 20:03:16 +08:00
  • 97b92b6b22 clean deep1 2023-04-22 20:02:32 +08:00
  • e4c04daadf git clone script working deep1 2023-04-22 18:42:01 +08:00
  • ba547ccb70 update download script deep1 2023-04-22 16:03:20 +08:00
  • 3410c9cad0 tidy wassname 2023-04-22 07:50:21 +00:00
  • d149f12704 readme wassname 2023-04-22 00:27:30 +00:00
  • 830c8f2605 gitkeep wassname 2023-04-22 00:26:42 +00:00
  • 2e7bf22d4a misc wassname 2023-04-22 00:24:57 +00:00
  • 4155cbb641 delora and test works :) wassname 2023-04-15 13:11:15 +00:00
  • e6338b5b26 rm int4 wassname 2023-04-15 06:37:43 +00:00
  • c520ad64ba trying to load int4 load_int4 wassname 2023-04-15 06:23:33 +00:00
  • 3a8c7f3a5b convert running, test working wassname 2023-04-10 19:47:53 +08:00
  • 039af1a0db readme wassname 2023-04-10 16:21:01 +08:00
  • e6f36838fe readme wassname 2023-04-10 16:17:13 +08:00
  • 6e6c8f8e7a init wassname 2023-04-10 16:15:52 +08:00
  • 5d3267d80d add v1 model as default in custom monkey patch John Smith 2023-04-10 09:33:41 +08:00
  • f185b90c3e Update README.md John Smith 2023-04-09 12:50:49 +08:00
  • 4c18a56fc0 fix bug John Smith 2023-04-09 12:44:50 +08:00
  • 8cf3bd4086 add g_idx support on cuda backend John Smith 2023-04-09 12:26:22 +08:00
  • b73f4e5e64 Merge pull request #64 from andybarry/readme_fix John Smith 2023-04-09 11:15:23 +08:00
  • b5d49cb9b1 Fix URL in readme. Andy Barry 2023-04-08 12:38:45 -04:00
  • 132c67be0d Fix bug John Smith 2023-04-08 23:58:30 +08:00
  • 56e5bf2854 Merge pull request #63 from andybarry/dockerfile John Smith 2023-04-08 15:48:27 +08:00
  • a93cf1264a Add timing on readme, remove useless line in dockerfile. Andy Barry 2023-04-08 01:54:29 -04:00
  • 191d92c940 Clean up diff Andy Barry 2023-04-08 01:27:56 -04:00
  • 31614fc2c4 Move 7bn changes into dockerfile. Andy Barry 2023-04-08 01:21:17 -04:00
  • 2e5aaf6dd6 Merge readmes. Andy Barry 2023-04-08 01:14:54 -04:00
  • e854f5d111 Fix after merge. Andy Barry 2023-04-08 00:53:28 -04:00
  • 8435b2c7f2 Merge branch 'main' of https://github.com/johnsmith0031/alpaca_lora_4bit Andy Barry 2023-04-07 22:02:54 -04:00
  • f91d4cbb59 Update README.md John Smith 2023-04-07 16:10:36 +08:00
  • b01b10eb4d Colorized output John Smith 2023-04-07 15:58:38 +08:00
  • 32904da1ff fix bug on triton matmul John Smith 2023-04-07 15:50:55 +08:00
  • dba3773b30 add triton backend support for v2 model John Smith 2023-04-07 15:34:06 +08:00
  • 9351f49542 merge pull request in new branch John Smith 2023-04-07 10:40:24 +08:00
  • 85e9cf004a Merge pull request #59 from yamashi/main John Smith 2023-04-07 10:05:18 +08:00
  • c5aa7fb695 Update finetune.py yamashi 2023-04-07 00:43:36 +02:00
  • 95cd390d25 Update Finetune4bConfig.py yamashi 2023-04-07 00:43:15 +02:00
  • 778035152d Update arg_parser.py yamashi 2023-04-07 00:42:34 +02:00
  • 2bf5d42f28 Add position_ids to flash attention yamashi 2023-04-06 17:46:15 +02:00
  • 7770e76c9c Fix args of flash attention yamashi 2023-04-06 17:32:01 +02:00
  • 30bf938d03 Update requirements.txt yamashi 2023-04-06 13:50:25 +02:00
  • 7b18b39dd8 Create llama_flash_attn_monkey_patch.py yamashi 2023-04-06 13:49:36 +02:00
  • 3ea18575c7 Use flash attention monkeypatch yamashi 2023-04-06 13:49:12 +02:00
  • 8020b3ec3b Update README.md John Smith 2023-04-06 13:57:32 +08:00
  • 9a02a88fb8 add patch for encode function to remove eos token at the beginning of left side John Smith 2023-04-06 12:56:27 +08:00
  • 3a95ad894b Update README.md Andy Barry 2023-04-06 00:53:57 -04:00
  • 1b8b6e193c Merge branch 'main' of github.com:andybarry/alpaca_lora_4bit_docker Andy Barry 2023-04-06 00:31:29 -04:00
  • 3076eaf4aa Add gif. Andy Barry 2023-04-06 00:30:28 -04:00
  • 0a51b6b0ee Update README.md Andy Barry 2023-04-06 00:20:30 -04:00
  • 6130b9bd0f Update README.md Andy Barry 2023-04-06 00:06:29 -04:00
  • 0f25304184 Update README.md Andy Barry 2023-04-06 00:06:05 -04:00
  • 242fc92e54 Fix url Andy Barry 2023-04-05 23:38:28 -04:00
  • fa966018f4 Fix docker run command. Andy Barry 2023-04-05 23:38:00 -04:00
  • 1af4023883 Fix some issues. Andy Barry 2023-04-05 23:29:10 -04:00
  • 417eba372a Add dockerfile and change some numbers to use 7bn model. Andy Barry 2023-04-05 23:13:35 -04:00
  • 085d9556f9 fix bug John Smith 2023-04-06 10:46:42 +08:00
  • c991e2a091 Delete triton_test.py Andrey Glushenkov 2023-04-06 02:39:40 +03:00
  • 0d271d5d90 Add files via upload Andrey Glushenkov 2023-04-06 02:38:06 +03:00
  • 4a2d23aa29 Delete autograd_4bit.py Andrey Glushenkov 2023-04-06 02:31:06 +03:00
  • f20570343f GPTQv2 support Andrey Glushenkov 2023-04-06 02:29:36 +03:00
  • 86387a0a35 update multi gpu support in finetune.py John Smith 2023-04-03 23:55:58 +08:00
  • 5655f218ed add g_idx buffer.\nadd triton matmul utils for future support. John Smith 2023-04-02 21:29:06 +08:00
  • f3a25342e1 fix device_map bug when using lora_apply_dir John Smith 2023-03-31 19:44:36 +08:00
  • 00bf0a1e1b Update README.md John Smith 2023-03-31 14:17:35 +08:00
  • dd0efc721f Merge pull request #47 from winglian/better-gpt4all John Smith 2023-03-31 11:20:03 +08:00
  • 8791eaee9a fix gpt4all training to more closely match the released logic, other small fixes and optimizations Wing Lian 2023-03-30 19:08:35 -04:00
  • 878eada8dd add amp_wrapper for autocast support. John Smith 2023-03-30 19:57:19 +08:00
  • b3c91a5af5 Merge pull request #45 from winglian/fix-missing-bracket John Smith 2023-03-30 13:53:55 +08:00
  • e744aec8bf fix missing paren Wing Lian 2023-03-29 23:40:30 -04:00
  • 8db4633d84 Update README.md John Smith 2023-03-30 11:24:25 +08:00
  • 8a62560e6c add offload support John Smith 2023-03-30 11:21:21 +08:00
  • 32976f91c4 Merge pull request #42 from winglian/multigpu-fix John Smith 2023-03-30 00:03:27 +08:00
  • b7361da58a better multi-gpu support, support gpt4all training data Wing Lian 2023-03-29 11:21:47 -04:00
  • 0fdae9224c optimized groupsize backward for performance John Smith 2023-03-29 17:44:51 +08:00
  • 5986649b37 Update README.md John Smith 2023-03-29 14:46:28 +08:00
  • 1c02d4262d add resume checkpoint to continue a training John Smith 2023-03-29 14:35:39 +08:00
  • 2a1cb42966 add padding support as an option John Smith 2023-03-29 11:20:16 +08:00
  • cff57ebfa4 Merge pull request #39 from winglian/fix-prompt-eos-token John Smith 2023-03-29 10:35:46 +08:00
  • daad59f8ef properly include the eos token so inference doesn't blabber on Wing Lian 2023-03-28 20:53:16 -04:00
  • 1719bd0ce3 fix bug John Smith 2023-03-29 08:09:40 +08:00
  • 1043ded7d9 Merge branch 'main' of github.com:johnsmith0031/alpaca_lora_4bit John Smith 2023-03-29 01:26:20 +08:00
  • d28ee06202 fix bug John Smith 2023-03-29 01:25:37 +08:00
  • b5e3dae573 Merge pull request #34 from winglian/v2-fixes John Smith 2023-03-28 23:49:56 +08:00
  • b47da33084 fixes for most recent update Wing Lian 2023-03-28 10:56:35 -04:00
  • 234004ceb5 fix bug John Smith 2023-03-28 22:05:18 +08:00
  • f26615fc0c fix bug John Smith 2023-03-28 21:47:22 +08:00
  • 0768d0fdff update finetune data format John Smith 2023-03-28 21:45:33 +08:00
  • 8a6c8661df Merge branch 'main' of github.com:johnsmith0031/alpaca_lora_4bit merged John Smith 2023-03-28 21:14:35 +08:00
  • 211af574b6 fix bug John Smith 2023-03-28 21:12:51 +08:00
  • ac07457473 Update README.md John Smith 2023-03-28 20:44:02 +08:00
  • bff039de95 add v2 model support John Smith 2023-03-28 20:33:55 +08:00
  • 667e43cb5b Merge pull request #30 from winglian/features/python-fixes John Smith 2023-03-28 09:34:50 +08:00
  • 101d314bd9 add missing dependency to train with LlamaTokenizer Wing Lian 2023-03-27 16:13:46 -04:00
  • 62e54ac1c7 backwards support for pre-py3.10, add datasets requirement used in train Wing Lian 2023-03-27 16:08:20 -04:00
  • 6c8c07e7ad Update README.md John Smith 2023-03-27 18:03:28 +08:00
  • cf94d7af68 Update README.md John Smith 2023-03-27 17:52:35 +08:00
  • 1ca9b8abf8 Update README.md John Smith 2023-03-27 17:51:04 +08:00
  • 0b5b376de1 Merge pull request #23 from sterlind/star/repos John Smith 2023-03-27 17:47:39 +08:00
  • 399c3d124e Tested and should be ready! Star Dorminey 2023-03-25 20:52:38 -07:00
  • a2a4c1d117 Remove gitmodules. Star Dorminey 2023-03-25 20:23:46 -07:00