admin/LLM-from-scratch
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • 6358aaa1ff Create README.md main Seeker 2026-01-08 13:47:20 +08:00
  • 17d0230f7e Update and rename README.md to README_zh.md Seeker 2026-01-08 13:46:38 +08:00
  • 671367eaa3 Update README description seeker 2025-12-15 10:51:14 +08:00
  • 1dc30fd0a8 Update tech.md IP127000 2025-08-06 15:40:08 +08:00
  • f9542b7c0c Create tech.md IP127000 2025-07-23 08:55:40 +08:00
  • d89fb5b876 Update README.md IP127000 2025-07-23 08:54:29 +08:00
  • 165d823891 [2025.06.22] add reward model PPO GRPO train code hanluzhi0429@gmail.com 2025-06-22 23:31:24 +08:00
  • 4a3c349d33 [2025.06.22] add reward model、PPO、GRPO train code hanluzhi0429@gmail.com 2025-06-22 22:57:51 +08:00
  • 2a0810012b [2025.06.22] add cold start hanluzhi0429@gmail.com 2025-06-22 22:12:48 +08:00
  • e4d1bf72b3 [2025.06.19] pretoken corpus and add cold start hanluzhi0429@gmail.com 2025-06-19 18:25:21 +08:00
  • aa0267da53 [2025.06.19] pre_token corpus and add cold start hanluzhi0429@gmail.com 2025-06-19 18:23:19 +08:00
  • bd75a6d3ac [2025.06.19] update train code, pretoken corpus root 2025-06-19 17:00:27 +08:00
  • f28b310bc3 [2025.06.17] update train code root 2025-06-17 18:24:05 +08:00
  • 5182ecc4a4 [2025.06.17] update train code root 2025-06-17 17:21:01 +08:00
  • 7ebb8a1b94 [2025.06.17] update train code root 2025-06-17 17:19:52 +08:00
  • ae76142fdc [2025.06.17] update train code root 2025-06-17 17:17:49 +08:00
  • c4f3a25653 [2025.06.13] remove __init__ files root 2025-06-13 15:40:28 +08:00
  • 9c1739baf3 [2025.06.10] train EN tokenizer use parquet files coffee 2025-06-10 17:56:00 +08:00
  • eb1c6c0c17 [2025.06.10]train en tokenizer root 2025-06-10 17:36:53 +08:00
  • c6e171e8ed [2025.06.10] adjust directory structure coffee 2025-06-10 14:04:57 +08:00
  • ef2f7a1bc5 [25.06.10] update code coffee 2025-06-10 13:58:12 +08:00
  • f9e0ae5825 [2025.06.06] update tokenizer train code coffee 2025-06-06 16:06:36 +08:00
  • 628aa8c674 [2025.06.05] add tokenzier test coffee 2025-06-05 18:35:39 +08:00
  • edc60be53b [2025.06.05] upload tokenizer weights and update the tokenizer train code coffee 2025-06-05 18:33:32 +08:00
  • dc4bbf43f6 [2025.06.05] add tokenzier test coffee 2025-06-05 18:20:36 +08:00
  • 0e023c8bc1 [2025.06.05] upload tokenizer weights and update the tokenizer train code coffee 2025-06-05 18:16:34 +08:00
  • 80517880b5 [2025.06.05] upload tokenizer weights and update the tokenizer train code coffee 2025-06-05 18:15:28 +08:00
  • 947ad2b87c [2025.06.05] upload tokenizer weights and update the tokenizer train code coffee 2025-06-05 18:12:48 +08:00
  • 6a30309848 [2025.06.05] update tokenizer train code, use the style of Qwen coffee 2025-06-05 18:09:18 +08:00
  • db4c2658b1 [2025.06.05] add BBPE tokenizer weights and update code coffee 2025-06-05 18:01:07 +08:00
  • b61572a8a1 [2025.06.04]update tokenizer train code and style coffee 2025-06-04 17:17:19 +08:00
  • 13c1edeb05 update tokenizer code, train from local files coffee 2025-06-04 17:15:09 +08:00
  • 319516e3fb rm llm coffee 2025-05-29 17:18:24 +08:00
  • d8c63e73aa add doc_tokenizer coffee 2025-05-29 17:16:48 +08:00
  • ba00ae0c36 [25.05.27] add unigramLM hanluzhi0429@gmail.com 2025-05-27 21:41:34 +08:00
  • 61b8b42b3c [25.05.26] add documents->tokenizer hanluzhi0429@gmail.com 2025-05-26 22:35:30 +08:00
  • 87fc8933ec [25.05.26] add documents->tokenizer hanluzhi0429@gmail.com 2025-05-26 22:31:22 +08:00
  • 6d7bdecac5 [25.05.26] add deepspeed stage3 config coffee 2025-05-26 17:40:32 +08:00
  • a7edf87dca [2025.05.22] add MoE support coffee 2025-05-22 18:13:01 +08:00
  • 9ae88672f5 [2025.05.22] add MoE support coffee 2025-05-22 18:11:50 +08:00
  • 6152ab5a35 add MoE support coffee 2025-05-22 17:52:24 +08:00
  • a229a11b0e [25.05.22] add MoE support coffee 2025-05-22 17:45:47 +08:00
  • 28fe25780e [2025.05.22] add MoE support coffee 2025-05-22 16:47:52 +08:00
  • 1d698ba7c3 add jsonl support coffee 2025-05-20 17:19:09 +08:00
  • b9b1d4b51d add jsonl support coffee 2025-05-20 17:17:49 +08:00
  • b262507cb8 [25.05.20] update code support resume from checkpoints coffee 2025-05-20 11:31:35 +08:00
  • 646c182469 update README.md coffee 2025-05-19 18:01:12 +08:00
  • 4cf94f0d48 update requirements coffee 2025-05-19 17:59:24 +08:00
  • 82f7c41bdc update requirements_version file coffee 2025-05-19 17:53:53 +08:00
  • 11473a798f remove pycache coffee 2025-05-19 15:43:02 +08:00
  • 5ca9ee3803 remove pycache coffee 2025-05-19 15:43:02 +08:00
  • acdbde74ce add torchrun train script coffee 2025-05-19 11:52:16 +08:00
  • 010580981e add torchrun train script coffee 2025-05-19 11:52:16 +08:00
  • 98a71bac61 25.05.19: add ds train code coffee 2025-05-19 11:51:33 +08:00
  • be180dfb6a 25.05.19: add ds train code coffee 2025-05-19 11:51:33 +08:00
  • ddde74e187 25.05.19:update pretrain_transformers code coffee 2025-05-19 11:50:14 +08:00
  • c2d9b96acc 25.05.19:update pretrain_transformers code coffee 2025-05-19 11:50:14 +08:00
  • 0482d6559d update dolly config coffee 2025-05-19 09:12:33 +08:00
  • c36f0911a7 update dolly config coffee 2025-05-19 09:12:33 +08:00
  • 2db94bfa0f add pretrain code coffee 2025-05-19 08:37:39 +08:00
  • 1ee644e8ae add pretrain code coffee 2025-05-19 08:37:39 +08:00
  • 7e55276b43 25.05.15:update modeling_dolly code and code comments coffee 2025-05-15 10:51:57 +08:00
  • b48ff8a049 25.05.15:update modeling_dolly code and code comments coffee 2025-05-15 10:51:57 +08:00
  • d52f33aa4a 25.05.15:update modeling_dolly code and code comments coffee 2025-05-15 10:46:58 +08:00
  • 630d761f26 25.05.15:update modeling_dolly code and code comments coffee 2025-05-15 10:46:58 +08:00
  • 48019ee38a 25.05.14:update modeling_dolly code and code comments coffee 2025-05-14 17:37:20 +08:00
  • bf0f04bcb3 25.05.14:update modeling_dolly code and code comments coffee 2025-05-14 17:37:20 +08:00
  • 6a4abf930e add code comments coffee 2025-05-14 16:00:08 +08:00
  • 793f8c8ea6 add code comments coffee 2025-05-14 16:00:08 +08:00
  • bdb1f0c2c9 update READEME.md coffee 2025-05-14 13:46:10 +08:00
  • 737b9dce04 update READEME.md coffee 2025-05-14 13:46:10 +08:00
  • 970ba259b8 update REAMDME.md coffee 2025-05-14 13:42:44 +08:00
  • 69025ca9a4 update REAMDME.md coffee 2025-05-14 13:42:44 +08:00
  • 2b19c0b1b3 add setup file to install dolly_llm coffee 2025-05-14 13:39:27 +08:00
  • b8022cc6b8 add setup file to install dolly_llm coffee 2025-05-14 13:39:27 +08:00
  • 165f61b8b1 add __init__ file coffee 2025-05-14 13:38:55 +08:00
  • 8e4f9e5de7 add __init__ file coffee 2025-05-14 13:38:55 +08:00
  • 6d2f929b51 update code coffee 2025-05-14 13:35:24 +08:00
  • 9d9e858b65 update code coffee 2025-05-14 13:35:24 +08:00
  • 8d9fe41b1b fix compute_dolly_parameters name coffee 2025-05-13 17:39:00 +08:00
  • e25a677be9 fix compute_dolly_parameters name coffee 2025-05-13 17:39:00 +08:00
  • a1000bceab add copute_dolly_parameters code coffee 2025-05-13 17:37:44 +08:00
  • cfedc4c1f8 add copute_dolly_parameters code coffee 2025-05-13 17:37:44 +08:00
  • 37685a5934 add compute paramters code coffee 2025-05-13 17:35:26 +08:00
  • f0583b43dd add compute paramters code coffee 2025-05-13 17:35:26 +08:00
  • 38dd484fc3 add test file coffee 2025-05-13 17:34:05 +08:00
  • fe280592bd add test file coffee 2025-05-13 17:34:05 +08:00
  • 892d3c47f2 update modeling_dolly code coffee 2025-05-13 17:33:33 +08:00
  • 575dcce05b update modeling_dolly code coffee 2025-05-13 17:33:33 +08:00
  • 63948eefb0 del trans coffee 2025-05-09 18:58:34 +08:00
  • c537905e9b del trans coffee 2025-05-09 18:58:34 +08:00
  • fcf5bd2814 update trans code coffee 2025-05-09 18:58:19 +08:00
  • e7cd5d1c23 update trans code coffee 2025-05-09 18:58:19 +08:00
  • a379580943 del trans coffee 2025-05-09 18:57:34 +08:00
  • 3312085568 del trans coffee 2025-05-09 18:57:34 +08:00
  • 4f75ee6021 update trans code coffee 2025-05-09 18:56:08 +08:00
  • 51d56faf9f update trans code coffee 2025-05-09 18:56:08 +08:00
  • 50a390dd26 update README.md coffee 2025-05-09 18:42:26 +08:00
  • 6c423fa4f8 update README.md coffee 2025-05-09 18:42:26 +08:00
  • 6684b4396e update README.md coffee 2025-05-09 18:31:23 +08:00