mirror of
https://github.com/LCTT/TranslateProject.git
synced 2024-12-26 21:30:55 +08:00
Merge pull request #28997 from lxbwolf/52-Open-Source-Model-Dolly-Claims-to-be-a-Cheaper-Alternative-to-ChatGPT
TSL
This commit is contained in:
Xingyu.Wang
GitHub
commit
06c4a9fafa
@ -1,83 +0,0 @@
|
|||||||
[#]: subject: "Open-Source Model 'Dolly' Claims to be a Cheaper Alternative to ChatGPT"
|
|
||||||
[#]: via: "https://news.itsfoss.com/open-source-model-dolly/"
|
|
||||||
[#]: author: "Sourav Rudra https://news.itsfoss.com/author/sourav/"
|
|
||||||
[#]: collector: "lkxed"
|
|
||||||
[#]: translator: "lxbwolf"
|
|
||||||
[#]: reviewer: " "
|
|
||||||
[#]: publisher: " "
|
|
||||||
[#]: url: " "
|
|
||||||
|
|
||||||
Open-Source Model 'Dolly' Claims to be a Cheaper Alternative to ChatGPT
|
|
||||||
======
|
|
||||||
|
|
||||||
An affordable alternative to ChatGPT? And, open-source? Looks like we're joining the open-source race against ChatGPT.
|
|
||||||
|
|
||||||
![open source model dolly][1]
|
|
||||||
|
|
||||||
![][2]
|
|
||||||
|
|
||||||
Databricks is a software company that has established itself in a variety of sectors, with data warehousing, and AI-based solutions being their primary focus.
|
|
||||||
|
|
||||||
In recent times, we have seen the meteoric rise of ChatGPT, resulting in similar efforts from the likes of Meta, Google, and even Mozilla.
|
|
||||||
|
|
||||||
And now, Databricks is trying in their own way by open-sourcing its [large language model][3] (LLM) 'Dolly'.
|
|
||||||
|
|
||||||
Let's take a look at it.
|
|
||||||
|
|
||||||
**What is happening?:** In a recent announcement, Databricks introduced what they term as **'a cheap-to-build'** LLM that functions by using an existing open-source parameter [model][4] by [EleutherAI][5].
|
|
||||||
|
|
||||||
The model has been slightly tweaked to give Dolly instruction following capabilities such as brainstorming and text generation.
|
|
||||||
|
|
||||||
When you compare the **175 billion parameters** in GPT-3, Dolly's **6 billion parameters** might seem puny in comparison.
|
|
||||||
|
|
||||||
But, the folks over at Databricks were surprised when they saw that even with this much data, Dolly was **able to****exhibit many of the same capabilities as ChatGPT**.
|
|
||||||
|
|
||||||
Below is one of the examples they showcased:
|
|
||||||
|
|
||||||
![a screenshot of how dolly performs in an open question and answer scenario][6]
|
|
||||||
|
|
||||||
The original model used data from [Alpaca][7], the model built by Stanford using the [LLaMA][8] LLM by Meta as a base.
|
|
||||||
|
|
||||||
But, as you can see, the original model produced a very haphazard result, whereas Dolly, with its different model and tweaks, was able to produce a far usable answer.
|
|
||||||
|
|
||||||
> 📝 Fun Fact: The name was taken from the first cloned mammal, Dolly the sheep.
|
|
||||||
|
|
||||||
**Why now?:** According to Databricks, they think that **many companies would prefer to build their own model** rather than sending data to some centralized provider who has locked their model behind an API.
|
|
||||||
|
|
||||||
Many companies might not be comfortable handing over their most sensitive data to a third party, and then there are the various tradeoffs in terms of model quality, cost, and desired behavior.
|
|
||||||
|
|
||||||
**Do you want to check it out?**
|
|
||||||
|
|
||||||
Sure, but there's a catch.
|
|
||||||
|
|
||||||
You will have to **use their platform to use Dolly**, they have open-sourced a [Databricks notebook][9] that will help you build it on Databricks.
|
|
||||||
|
|
||||||
Moreover, if you want to get access to the trained weights, you will have to contact them. I am uncertain whether they will provide access to it for free, though.
|
|
||||||
|
|
||||||
In a nutshell, this move to open-source their model should be good for companies to help safeguard their data, save on operating costs, and more by enabling them to create their own model.
|
|
||||||
|
|
||||||
You can check out the[announcement blog][10] to learn more about the technical details and other plans for it.
|
|
||||||
|
|
||||||
--------------------------------------------------------------------------------
|
|
||||||
|
|
||||||
via: https://news.itsfoss.com/open-source-model-dolly/
|
|
||||||
|
|
||||||
作者：[Sourav Rudra][a]
|
|
||||||
选题：[lkxed][b]
|
|
||||||
译者：[译者ID](https://github.com/译者ID)
|
|
||||||
校对：[校对者ID](https://github.com/校对者ID)
|
|
||||||
|
|
||||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
|
||||||
|
|
||||||
[a]: https://news.itsfoss.com/author/sourav/
|
|
||||||
[b]: https://github.com/lkxed/
|
|
||||||
[1]: https://news.itsfoss.com/content/images/size/w1304/2023/03/opensource-ai-model-dolly.png
|
|
||||||
[2]: https://news.itsfoss.com/content/images/2023/03/linux-mega-packt.webp
|
|
||||||
[3]: https://en.wikipedia.org/wiki/Large_language_model?ref=its-foss-news
|
|
||||||
[4]: https://huggingface.co/EleutherAI/gpt-j-6B?ref=its-foss-news
|
|
||||||
[5]: https://www.eleuther.ai/?ref=its-foss-news
|
|
||||||
[6]: https://news.itsfoss.com/content/images/2023/03/Dolly_AI.jpg
|
|
||||||
[7]: https://crfm.stanford.edu/2023/03/13/alpaca.html?ref=its-foss-news
|
|
||||||
[8]: https://ai.facebook.com/blog/large-language-model-llama-meta-ai/?ref=its-foss-news
|
|
||||||
[9]: https://github.com/databrickslabs/dolly?ref=its-foss-news
|
|
||||||
[10]: https://www.databricks.com/blog/2023/03/24/hello-dolly-democratizing-magic-chatgpt-open-models.html?ref=its-foss-news
|
|
||||||
@ -0,0 +1,83 @@
|
|||||||
|
[#]: subject: "Open-Source Model 'Dolly' Claims to be a Cheaper Alternative to ChatGPT"
|
||||||
|
[#]: via: "https://news.itsfoss.com/open-source-model-dolly/"
|
||||||
|
[#]: author: "Sourav Rudra https://news.itsfoss.com/author/sourav/"
|
||||||
|
[#]: collector: "lkxed"
|
||||||
|
[#]: translator: "lxbwolf"
|
||||||
|
[#]: reviewer: " "
|
||||||
|
[#]: publisher: " "
|
||||||
|
[#]: url: " "
|
||||||
|
|
||||||
|
号称可以成为 ChatGPT 平替的开源模型 ”Dolly“
|
||||||
|
======
|
||||||
|
|
||||||
|
你需要一款 ChatGPT 的平替？还得是开源的？看起来我们已经被卷入了与 ChatGPT 的开源大战。
|
||||||
|
|
||||||
|
![open source model dolly][1]
|
||||||
|
|
||||||
|
![][2]
|
||||||
|
|
||||||
|
Databricks 这家软件公司，在各个领域都有所建树，尤其是在数据仓库和基于人工智能的解决方案方面。
|
||||||
|
|
||||||
|
最近，随着 ChatGPT 横空出世，Meta、Google 甚至 Mozilla 也可以效仿 ChatGPT。
|
||||||
|
|
||||||
|
而现在，Databricks 开源了其 [<ruby>大型语言模型<rt>large language model</rt></ruby>][3]（LLM）"Dolly"，也正在以自己的方式进行尝试。
|
||||||
|
|
||||||
|
我们一起来看看它。
|
||||||
|
|
||||||
|
**发生了什么？：** 在最近的公告中，Databricks 介绍了他们号称 **”廉价构建“** 的 LLM，使用 [EleutherAI][5] 的已经开源参数[模型][4] 提供功能。
|
||||||
|
|
||||||
|
他们在该模型基础上稍作调整，赋予了 Dolly 指令跟随能力，如头脑风暴和文本生成。
|
||||||
|
|
||||||
|
当你拿它与 GPT-3 中的 **1750亿个参数** 比较时，Dolly 的 **60亿个参数** 就可能显得微不足道了。
|
||||||
|
|
||||||
|
但是，当 Databricks 的人看到即使数据量与 GPT-3 相差这么多，Dolly 也能 **展示很多与 ChatGPT 相同的能力** 时，他们感到非常震惊。
|
||||||
|
|
||||||
|
下面是他们展示的其中一个例子：
|
||||||
|
|
||||||
|
![a screenshot of how dolly performs in an open question and answer scenario][6]
|
||||||
|
|
||||||
|
原始模型使用了 [Alpaca][7] 的数据，该模型由斯坦福大学以 Meta 的 [LaMA][8] LLM 为基础建立。
|
||||||
|
|
||||||
|
但是，正如你所看到的，原始模型产生了一个非常杂乱无章的结果，而 Dolly，通过不同的模型和调整，能够产生一个更为可用的答案。
|
||||||
|
|
||||||
|
> 📝 有趣的事实：Dolly 名字取自世界上第一只克隆羊。
|
||||||
|
|
||||||
|
**为什么是现在？** 根据 Databricks 的说法，他们认为 **许多公司更愿意建立自己的模型，**而不是将数据发送给某个紧紧掌握模型只对外提供 API 的集中式供应商。
|
||||||
|
|
||||||
|
许多公司可能不愿意将他们最敏感的数据交给第三方，然后在模型质量、成本和所需行为方面进行各种权衡。
|
||||||
|
|
||||||
|
**你想看看吗？**
|
||||||
|
|
||||||
|
当然，但有一个问题。
|
||||||
|
|
||||||
|
你必须 **使用他们的平台来使用 Dolly**，他们已经开源了一个 [Databricks 笔记本][9]，可以帮助你在 Databricks 上构建它。
|
||||||
|
|
||||||
|
此外，如果你想获得训练好的权重，你必须联系他们。不过我不确定他们是否会免费提供使用权。
|
||||||
|
|
||||||
|
总而言之，这种开源其模型的举动应该对其他公司有好处，可以保护他们的数据、节省运营成本，其他公司也能使用它创建自己的模型。
|
||||||
|
|
||||||
|
你可以查看 [公告博客][10]，以了解更多技术细节和其他计划。
|
||||||
|
|
||||||
|
--------------------------------------------------------------------------------
|
||||||
|
|
||||||
|
via: https://news.itsfoss.com/open-source-model-dolly/
|
||||||
|
|
||||||
|
作者：[Sourav Rudra][a]
|
||||||
|
选题：[lkxed][b]
|
||||||
|
译者：[lxbwolf](https://github.com/lxbwolf)
|
||||||
|
校对：[校对者ID](https://github.com/校对者ID)
|
||||||
|
|
||||||
|
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||||
|
|
||||||
|
[a]: https://news.itsfoss.com/author/sourav/
|
||||||
|
[b]: https://github.com/lkxed/
|
||||||
|
[1]: https://news.itsfoss.com/content/images/size/w1304/2023/03/opensource-ai-model-dolly.png
|
||||||
|
[2]: https://news.itsfoss.com/content/images/2023/03/linux-mega-packt.webp
|
||||||
|
[3]: https://en.wikipedia.org/wiki/Large_language_model?ref=its-foss-news
|
||||||
|
[4]: https://huggingface.co/EleutherAI/gpt-j-6B?ref=its-foss-news
|
||||||
|
[5]: https://www.eleuther.ai/?ref=its-foss-news
|
||||||
|
[6]: https://news.itsfoss.com/content/images/2023/03/Dolly_AI.jpg
|
||||||
|
[7]: https://crfm.stanford.edu/2023/03/13/alpaca.html?ref=its-foss-news
|
||||||
|
[8]: https://ai.facebook.com/blog/large-language-model-llama-meta-ai/?ref=its-foss-news
|
||||||
|
[9]: https://github.com/databrickslabs/dolly?ref=its-foss-news
|
||||||
|
[10]: https://www.databricks.com/blog/2023/03/24/hello-dolly-democratizing-magic-chatgpt-open-models.html?ref=its-foss-news
|
||||||
Loading…
Reference in New Issue
Block a user