gnh1201/Open-Llama
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update readme

Browse Source
This commit is contained in:
LiangSong 2023-03-29 21:51:51 +08:00
parent a804d7d391
commit d065e3943c
2 changed files with 4 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
README.md
View File

@ -17,6 +17,7 @@ Open-Llama是一个开源项目，提供了一整套用于构建大型语言模
## 进展 ## 进展
经过30K step的预训练，已经展现出了一定的胡说八道的能力，如下分别是编写代码、续写论文和续写中文，虽然正确性存在问题但是已经像是一句话了。 经过30K step的预训练，已经展现出了一定的胡说八道的能力，如下分别是编写代码、续写论文和续写中文，虽然正确性存在问题但是已经像是一句话了。
<img src="assets/code.JPG" width="33%"><img src="assets/paper.JPG" width="33%"><img src="assets/chinese.JPG" width="33%"> <img src="assets/code.JPG" width="33%"><img src="assets/paper.JPG" width="33%"><img src="assets/chinese.JPG" width="33%">
## **特性** ## **特性**

10
README_en.md
View File

@ -2,7 +2,7 @@
* @Author: LiangSong(sl12160010@gmail.com) * @Author: LiangSong(sl12160010@gmail.com)
* @Date: 2023-03-10 21:18:35 * @Date: 2023-03-10 21:18:35
* @LastEditors: LiangSong(sl12160010@gmail.com) * @LastEditors: LiangSong(sl12160010@gmail.com)
* @LastEditTime: 2023-03-29 21:44:43 * @LastEditTime: 2023-03-29 21:51:28
* @FilePath: /Open-Llama/README_en.md * @FilePath: /Open-Llama/README_en.md
* @Description: * @Description:
* *
@ -15,13 +15,9 @@ Translated by ChatGPT.
Open-Llama is an open source project that provides a complete set of training processes for building large-scale language models, from data preparation to tokenization, pre-training, instruction tuning, and reinforcement learning techniques such as RLHF. Open-Llama is an open source project that provides a complete set of training processes for building large-scale language models, from data preparation to tokenization, pre-training, instruction tuning, and reinforcement learning techniques such as RLHF.
## Progress ## Progress
After 30K steps of pre-training, the model has demonstrated some language capabilities, as shown below in coding, paper continuation, and Chinese continuation tasks. Although there are still issues with correctness, the generated outputs resemble sentences. After 30K steps of pre-training, the model has demonstrated some language capabilities, as shown below in coding and paper continuation tasks. Although there are still issues with correctness, the generated outputs resemble sentences.
<div style="display:flex;"> <img src="assets/code.JPG" width="33%"><img src="assets/paper.JPG" width="50%"><img src="assets/paper.JPG" width="50%">
<div style="flex:1"><img src="assets/code.JPG" width="100%"></div>
<div style="flex:1"><img src="assets/paper.JPG" width="100%"></div>
<div style="flex:1"><img src="assets/chinese.JPG" width="100%"></div>
</div>
## **Features** ## **Features**
### Ease of Use ### Ease of Use