Commit Graph

181 Commits

Author SHA1 Message Date
coffee 9c1739baf3 [2025.06.10] train EN tokenizer use parquet files 2025-06-10 17:56:00 +08:00
root eb1c6c0c17 [2025.06.10]train en tokenizer 2025-06-10 17:36:53 +08:00
coffee c6e171e8ed [2025.06.10] adjust directory structure 2025-06-10 14:04:57 +08:00
coffee ef2f7a1bc5 [25.06.10] update code 2025-06-10 13:58:12 +08:00
coffee f9e0ae5825 [2025.06.06] update tokenizer train code 2025-06-06 16:06:36 +08:00
coffee 628aa8c674 [2025.06.05] add tokenzier test 2025-06-05 18:35:39 +08:00
coffee edc60be53b [2025.06.05] upload tokenizer weights and update the tokenizer train code 2025-06-05 18:33:32 +08:00
coffee dc4bbf43f6 [2025.06.05] add tokenzier test 2025-06-05 18:20:36 +08:00
coffee 0e023c8bc1 [2025.06.05] upload tokenizer weights and update the tokenizer train code 2025-06-05 18:16:34 +08:00
coffee 80517880b5 [2025.06.05] upload tokenizer weights and update the tokenizer train code 2025-06-05 18:15:28 +08:00
coffee 947ad2b87c [2025.06.05] upload tokenizer weights and update the tokenizer train code 2025-06-05 18:12:48 +08:00
coffee 6a30309848 [2025.06.05] update tokenizer train code, use the style of Qwen 2025-06-05 18:09:18 +08:00
coffee db4c2658b1 [2025.06.05] add BBPE tokenizer weights and update code 2025-06-05 18:01:07 +08:00
coffee b61572a8a1 [2025.06.04]update tokenizer train code and style 2025-06-04 17:17:19 +08:00
coffee 13c1edeb05 update tokenizer code, train from local files 2025-06-04 17:15:09 +08:00
coffee 319516e3fb rm llm 2025-05-29 17:18:24 +08:00
coffee d8c63e73aa add doc_tokenizer 2025-05-29 17:16:48 +08:00
hanluzhi0429@gmail.com ba00ae0c36 [25.05.27] add unigramLM 2025-05-27 21:41:34 +08:00
hanluzhi0429@gmail.com 61b8b42b3c [25.05.26] add documents->tokenizer 2025-05-26 22:35:30 +08:00
hanluzhi0429@gmail.com 87fc8933ec [25.05.26] add documents->tokenizer 2025-05-26 22:31:22 +08:00
coffee 6d7bdecac5 [25.05.26] add deepspeed stage3 config 2025-05-26 17:40:32 +08:00
coffee a7edf87dca [2025.05.22] add MoE support 2025-05-22 18:13:01 +08:00
coffee 9ae88672f5 [2025.05.22] add MoE support 2025-05-22 18:11:50 +08:00
coffee 6152ab5a35 add MoE support 2025-05-22 17:52:24 +08:00
coffee a229a11b0e [25.05.22] add MoE support 2025-05-22 17:45:47 +08:00
coffee 28fe25780e [2025.05.22] add MoE support 2025-05-22 16:47:52 +08:00
coffee 1d698ba7c3 add jsonl support 2025-05-20 17:19:09 +08:00
coffee b9b1d4b51d add jsonl support 2025-05-20 17:17:49 +08:00
coffee b262507cb8 [25.05.20] update code support resume from checkpoints 2025-05-20 11:31:35 +08:00
coffee 646c182469 update README.md 2025-05-19 18:01:12 +08:00
coffee 4cf94f0d48 update requirements 2025-05-19 17:59:24 +08:00
coffee 82f7c41bdc update requirements_version file 2025-05-19 17:53:53 +08:00
coffee 11473a798f remove pycache 2025-05-19 15:43:02 +08:00
coffee 5ca9ee3803 remove pycache 2025-05-19 15:43:02 +08:00
coffee acdbde74ce add torchrun train script 2025-05-19 11:52:16 +08:00
coffee 010580981e add torchrun train script 2025-05-19 11:52:16 +08:00
coffee 98a71bac61 25.05.19: add ds train code 2025-05-19 11:51:33 +08:00
coffee be180dfb6a 25.05.19: add ds train code 2025-05-19 11:51:33 +08:00
coffee ddde74e187 25.05.19:update pretrain_transformers code 2025-05-19 11:50:14 +08:00
coffee c2d9b96acc 25.05.19:update pretrain_transformers code 2025-05-19 11:50:14 +08:00
coffee 0482d6559d update dolly config 2025-05-19 09:12:33 +08:00
coffee c36f0911a7 update dolly config 2025-05-19 09:12:33 +08:00
coffee 2db94bfa0f add pretrain code 2025-05-19 08:46:46 +08:00
coffee 1ee644e8ae add pretrain code 2025-05-19 08:46:46 +08:00
coffee 7e55276b43 25.05.15:update modeling_dolly code and code comments 2025-05-15 10:51:57 +08:00
coffee b48ff8a049 25.05.15:update modeling_dolly code and code comments 2025-05-15 10:51:57 +08:00
coffee d52f33aa4a 25.05.15:update modeling_dolly code and code comments 2025-05-15 10:46:58 +08:00
coffee 630d761f26 25.05.15:update modeling_dolly code and code comments 2025-05-15 10:46:58 +08:00
coffee 48019ee38a 25.05.14:update modeling_dolly code and code comments 2025-05-14 17:37:20 +08:00
coffee bf0f04bcb3 25.05.14:update modeling_dolly code and code comments 2025-05-14 17:37:20 +08:00