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 2023-04-09 22:48:56 +08:00
parent 00cda9e265
commit f9e7a3376a
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -2,7 +2,7 @@
* @Author: LiangSong(sl12160010@gmail.com)
* @Date: 2023-03-10 21:18:35
* @LastEditors: LiangSong(sl12160010@gmail.com)
* @LastEditTime: 2023-04-07 23:19:21
* @LastEditTime: 2023-04-09 22:48:28
* @FilePath: /Open-Llama/README.md
* @Description:
*
@ -16,6 +16,8 @@ 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
**采用FastChat项目相同方法测评Open-Llama的效果和GPT3.5的效果对比，经过测试在中文问题上可以达到GPT3.5 84%的水平，具体测试结果和CheckPoint将在近期放出**
我们完成了300B token的预训练，总共训练80 K step，Global Batch Size和Llama中一致为4M。
使用总共7部分数据构成Instruction-tuning数据，模型具有一定的编程能力、数学能力和多轮对话能力，具体数据见Instruction-Tuning部分。