Git Product home page Git Product logo

Comments (6)

formath avatar formath commented on May 20, 2024

推算一下,7B模型,1.2万亿token,1000张A800,0.58利用率,训练一个epoch是4天左右。

from baichuan-7b.

miraclezqc avatar miraclezqc commented on May 20, 2024

推算一下,7B模型,1.2万亿token,1000张A800,0.58利用率,训练一个epoch是4天左右。

看配置好像是纯data parallel,没有开tensor parallel吗?

from baichuan-7b.

formath avatar formath commented on May 20, 2024

推算一下,7B模型,1.2万亿token,1000张A800,0.58利用率,训练一个epoch是4天左右。

看配置好像是纯data parallel,没有开tensor parallel吗?

猜测应该开了tensor和pipeline并行,否则很难达到0.58利用率

from baichuan-7b.

miraclezqc avatar miraclezqc commented on May 20, 2024

推算一下,7B模型,1.2万亿token,1000张A800,0.58利用率,训练一个epoch是4天左右。

看配置好像是纯data parallel,没有开tensor parallel吗?

猜测应该开了tensor和pipeline并行,否则很难达到0.58利用率

7B开pipeline应该不至于,tp开的话可能也是2,因为seq length为4096,按照global batch size为4M推测,micro batch size和 gradient accumulate都是1,那千卡应该是纯dp的,除非是2000卡。。

from baichuan-7b.

Luoyingfeng8 avatar Luoyingfeng8 commented on May 20, 2024

推算一下,7B模型,1.2万亿token,1000张A800,0.58利用率,训练一个epoch是4天左右。

看配置好像是纯data parallel,没有开tensor parallel吗?

猜测应该开了tensor和pipeline并行,否则很难达到0.58利用率

7B开pipeline应该不至于,tp开的话可能也是2,因为seq length为4096,按照global batch size为4M推测,micro batch size和 gradient accumulate都是1,那千卡应该是纯dp的,除非是2000卡。。

7B在80G上不用开TP,只需要sharding=8即可,多机间就是纯dp,这样训练速度和吞吐量应该都是最优的

from baichuan-7b.

mynewstart avatar mynewstart commented on May 20, 2024

我想问下这个代码是把数据一次性加载进内存了,如果数据量很大1.4T tokens大概5T左右的数据量,是不是内存放不下呀。

from baichuan-7b.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.