admin/LLM-from-scratch
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
14 Commits 1 Branch 0 Tags
3dfa1324966a75aff79504f3e049c9a89060ebbb
T
Clone
Open with VS Code Open with VSCodium Open with Intellij IDEA
Download ZIP Download TAR.GZ Download BUNDLE
coffee 3dfa132496 update README log
2025-04-29 18:44:48 +08:00
corpus
add corpus files and update ignore
2025-04-29 18:35:28 +08:00
tokenizer_code
add tokenizer_code readme
2025-04-29 18:33:35 +08:00
tokenizer_file
del some tokenizer model file
2025-04-29 18:24:43 +08:00
utils
add utils for data process
2025-04-29 18:28:45 +08:00
.gitignore
add corpus files and update ignore
2025-04-29 18:35:28 +08:00
LICENSE
Initial commit
2025-04-23 22:42:17 +08:00
README.md
update README log
2025-04-29 18:44:48 +08:00
requirements.txt
add tokenizer code
2025-04-29 18:21:17 +08:00
test_huggingface_train.py
test huggingface train
2025-04-24 22:32:31 +08:00
test_huggingface_transformer.py
test huggingface transformers
2025-04-24 22:19:55 +08:00
test.py
add tokenizer code
2025-04-29 18:21:17 +08:00

README.md

dolly

Dolly is a custom LLM built from scratch

更新日志

2025-04-29

  • 完成tokenizer构建代码，支持sentencepiece和transfomers方式，支持从文本构建和从已有的tokenzier构建。

致谢

特别感谢以下资源和文章的帮助：

Reference in New Issue View Git Blame Copy Permalink
S
Description
大语言模型（LLM）从零训练流程的最小复现。
Readme 7.8 MiB
Languages
Python 99.6%
Shell 0.4%