Giter VIP home page Giter VIP logo

Comments (14)

Liangdi avatar Liangdi commented on August 11, 2024 10

我测试了也是这样,没有做对话训练的结果把, 相当于其他公司的 base 模型,后期应该会有 chat 的模型

from yi.

LiuChen19960902 avatar LiuChen19960902 commented on August 11, 2024 7

这又不是chat模型

from yi.

mzh1996 avatar mzh1996 commented on August 11, 2024 6

这个应该是base版本模型,我这边测试发现也只是做补全,就像llama1刚放出来时那样

from yi.

White-Friday avatar White-Friday commented on August 11, 2024 3

这是base模型,不是chat模型,等chat模型公布在测试

from yi.

jiangchengSilent avatar jiangchengSilent commented on August 11, 2024 2

建议使用finetune代码训练之后的chat模型试一下

from yi.

YIZXIY avatar YIZXIY commented on August 11, 2024 2

这破模型,根本就不看类型,也不看自述文件,就在这胡编乱造,令人无语

from yi.

ZhaoFancy avatar ZhaoFancy commented on August 11, 2024 2

image 为什么我的提示是中文,就啥也没输出?

@lkp1985 单看截图,回复里多了个?。你可以把 eos_token='\n' 去掉对比看下。

from yi.

ZhaoFancy avatar ZhaoFancy commented on August 11, 2024 1

Chat 模型已经发布,可以再试试呢? 🤗

from yi.

xihajun avatar xihajun commented on August 11, 2024

这破模型,根本就不看类型,也不看自述文件,就在这胡编乱造,令人无语

可以自己finetune一下

from yi.

bot66 avatar bot66 commented on August 11, 2024

README写了是预训练模型,不懂的话可以先友好的问一下。

from yi.

lkp1985 avatar lkp1985 commented on August 11, 2024

image
为什么我的提示是中文,就啥也没输出?

from yi.

krisshen2021 avatar krisshen2021 commented on August 11, 2024

它可以用中文写instruction,按照alpaca的格式来写,并需要给到它context定义,使用它的GPTQ版本可以较好的进行支持对话,而且是uncensored的,对中文支持也比较好,但是经常会出现回复重复的问题,中文的某些资料不全哈哈哈。

from yi.

lkp1985 avatar lkp1985 commented on August 11, 2024

image 为什么我的提示是中文,就啥也没输出?

@lkp1985 单看截图,回复里多了个?。你可以把 eos_token='\n' 去掉对比看下。

找到问题了,是tokenizer的原因,把tokenizer=''就可以了

from yi.

xihajun avatar xihajun commented on August 11, 2024

它可以用中文写instruction,按照alpaca的格式来写,并需要给到它context定义,使用它的GPTQ版本可以较好的进行支持对话,而且是uncensored的,对中文支持也比较好,但是经常会出现回复重复的问题,中文的某些资料不全哈哈哈。

请问具体是如何做的呢?这样

[SYSTEM]{system message}[/SYSTEM]
[INST]{instructions}[/INST]
{response}

from yi.

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.