gpt-2 文章自动生成
意见 (Opinion)
I usually would post interesting articles related to the topics of Artificial Intelligence and Machine Learning that I have come across on Medium this week.
我通常会在本周在Medium上发布与人工智能和机器学习主题相关的有趣文章。
But recently, it’s rather hard to escape the hype and interest around GPT-3.
但是最近，要逃避对GPT-3的炒作和关注是相当困难的。
So this week I’ve identified a few articles that will take your knowledge of GPT-3 from zero to ‘kinda’ hero.
因此，本周我确定了一些文章，这些文章将使您对GPT-3的了解从零变为“ 有点 ”英雄。
The presented articles will provide you with some knowledge to understand the essential background of GPT-3, along with its intuitive applications and potential threats.
所介绍的文章将为您提供一些知识，以了解GPT-3的基本背景，其直观的应用程序和潜在的威胁。
For those who are somewhat in a hurry, feel free to only read the ‘Short Review’ of the summary and review of the presented articles.
对于那些有些着急的人，请随意阅读 摘要 的“ 简短审阅” 和对所提交文章的审阅。
And those with a bit more time can delve into the ‘Detailed Review’ of the summary to understand my thoughts and opinions of the mentioned articles.
那些时间比较多的人可以深入研究 摘要 的“ 详细审阅” ，以了解我对所提及文章的想法和看法。
您将找到文章： (You will find articles:)
That explains GPT-3 in 3 minutes
这说明了3分钟内的GPT-3
Or, explain why GPT-3 is overhyped, also in 3 minutes
或者，在3分钟内解释为什么GPT-3被过度炒作了
Provide an in-depth understanding of GPT-3
深入了解GPT-3
Question the longevity of coding roles due to GPT-3 potential.
质疑由于GPT-3潜力而产生的编码角色的寿命。
GPT-3：第一个人工智能？ 朱利安·劳雷特(Julien Lauret) (GPT-3: The First Artificial General Intelligence? By Julien Lauret)
One-stop-shop to get information into the history, development and potential of GPT-3.
一站式服务，以获取有关GPT-3的历史，发展和潜力的信息。
简短评论 (Short Review)
Julien Lauret’s article is a comprehensive summary of the journey taken so far to create GPT-3.
朱利安·劳雷特(Julien Lauret)的文章全面概述了迄今为止创建GPT-3的过程。
Julien has managed to summarize years of development and introductions of methodology and techniques to model language and solve natural language processing into several small, concise paragraphs.
朱利安(Julien)已成功总结了多年的发展以及对语言建模和解决自然语言处理问题的方法论和技术的介绍，并分为几个简短的小段。
As well as providing the reader with a background of GPT-3, Julien also gives a somewhat diplomatic answer to the question as to whether GPT-3 is AGI. His response truly reflects the nature of the question itself, in that the question is subjected to the definition of intelligence by whoever poses the question.
除了为读者提供GPT-3的背景知识外，Julien还对GPT-3是否为AGI提出了一些外交上的回答。 他的回答确实反映了问题本身的性质，因为提出问题的人都会对问题进行智力定义。
详细审查 (Detailed Review)
Julien’s latest article on GPT-3 is a well-needed article, as I feel the topic of Artificial General Intelligence(AGI) has almost taken a backseat in our imagination.
朱利安(Julien)在GPT-3上发表的最新文章是一篇急需的文章，因为我觉得人工智能(AGI)的话题几乎在我们的想象中倒退了。
Julien points early in the article, two viewpoints on AGI. One viewpoint alludes to the fact that AGI is decades away; while the other questions the possibility of humans ever achieving AGI.
朱利安(Julien)在文章开头指出了关于AGI的两种观点。 一种观点暗示着AGI距今已有数十年之久。 而其他人则质疑人类实现AGI的可能性。
Julien’s article is littered with some technical and specialist terms, but each term is accompanied with a brief definition and in some cases an extensive background.
朱利安(Julien)的文章中充斥着一些技术和专业术语，但每个术语都带有简短定义，在某些情况下还具有广泛的背景。
Terms such as ‘Deep Neural Networks’, ‘Machne translations’, ‘Word2vec’ and ‘Few-shot learning’ are presented and defined in a manner that provides readers with a crash course on language modelling and NLP, as well as the primary information around GPT-3.
呈现和定义“深度神经网络”，“ Machne翻译”，“ Word2vec”和“少量学习”等术语，为读者提供有关语言建模和NLP的速成课程以及主要信息大约是GPT-3。
The article doesn’t mention GPT-3 until halfway through the article, due to the fact that Julien cleverly takes the reader through a journey that outlines crucial developments and research that have contributed to the advancement of language modelling and GPT-3 itself.
由于朱利安(Julien)巧妙地引导读者完成概述语言开发和研究的旅程，这些进展为语言建模和GPT-3本身的发展做出了贡献，因此直到文章中途才提到GPT-3。
Information on the intrinsic characteristics(number of weights) of both GPT-3 and its predecessor are provided to give the reader an understanding of the level of progress the GPT model has made at every released version.
提供了有关GPT-3及其前身的内在特性(权重数量)的信息，以使读者了解GPT模型在每个发行版本中所取得的进步水平。
The latter half of the article presents examples of the intuitive application of GPT-3, developed by early access API users.
本文的后半部分展示了由早期访问API用户开发的GPT-3直观应用程序的示例。
But more importantly, Julien answers the question proposed in the article’s title. Julien’s conclusion at to whether if GPT-3 is AGI falls on the conclusion that it isn’. But even more so, the answer to the question itself is not as clear as one might think.
但更重要的是，朱利安回答了文章标题中提出的问题。 朱利安关于GPT-3是否为AGI的结论基于“它不是”的结论。 但更重要的是，问题本身的答案并不像人们想象的那么清晰。
The unclarity stems from the fact that the definition of intelligence is ambiguous and subject to an individual’s interpretation.
不确定性源于这样一个事实，即智力的定义是模棱两可的，并受个人的解释。
What Julien states that is an important matter to discuss is the application of GPT-3. We’ve already seen its ability to generate poems, gameplay scenarios and web components defined through language.
朱利安指出，讨论的重要事项是GPT-3的应用。 我们已经看到了其生成诗歌，游戏场景和通过语言定义的Web组件的能力。
What we now need to observe is how the latest feat of AI can be applied in a more general setting that involves videos and images.
我们现在需要观察的是，如何在涉及视频和图像的更一般的设置中应用最新的AI专长。
Julien’s article interestingly contains segments of philosophical views and statements. For example, Julien points to the proposition that we can’t determine the level of intelligence GPT-3 possesses, just as we can’t determine the level of cognitive function of a paralyzed person or individual with physical limitations.
朱利安的文章有趣地包含了一些哲学观点和陈述。 例如，朱利安(Julien)指出我们无法确定GPT-3拥有的智力水平，就像我们无法确定具有身体限制的瘫痪者或个人的认知功能水平一样。
This article delves touches areas of AI that are worth pondering about.
本文深入探讨了值得深思的AI领域。
这篇文章非常适合阅读： (This article is a great read for:)
Machine Learning Practitioners: The language and technical term used is very familiar with anyone involved in machine learning. And the definitions and papers included allows for further exploitation of topics related to GPT-3
机器学习从业者 ：所使用的语言和技术术语对参与机器学习的任何人都非常熟悉。 并且所包含的定义和论文可以进一步利用与GPT-3相关的主题
GPT-3在3分钟内得到了解释，作者Dale Markowitz (GPT-3 Explained In Under 3 Minutes By Dale Markowitz)
Short bite-size information on GPT-3
有关GPT-3的简短信息
简短评论 (Short Review)
For those who are slightly short of time and would like to get bite-size information on GPT-3, then Dale Markowitz’s article is the perfect stop. With well structured and presented paragraphs, you get the important information on key properties of GPT-3. The article also includes some examples of GPT-3 based applications.
对于那些时间比较短并且想要了解有关GPT-3的信息的人， Dale Markowitz的文章是理想的选择。 通过精心组织和呈现的段落，您可以获得有关GPT-3关键属性的重要信息。 本文还包括一些基于GPT-3的应用程序的示例。
详细审查 (Detailed Review)
Dale’s article can be described as a quick shot of useful information that contains the important details on GPT-3.
Dale的文章可以描述为有用信息的简要介绍，其中包含有关GPT-3的重要细节。
Dale included in the article an exploration into the general makeup of the GPT-3 language model without the extensive inclusion of technical jargons and details.
Dale在本文中探讨了GPT-3语言模型的一般构成，但没有广泛包含技术术语和细节。
The article starts with some showcases of GPT-3 based applications through a series of embedded twitter posts.
本文首先通过一系列嵌入式Twitter帖子展示了一些基于GPT-3的应用程序。
Dale’s article provides information into the definition of the common technical idea GPT-3 is associated with, which is ‘language model’.
Dale的文章为GPT-3与之相关的通用技术思想(即“语言模型”)的定义提供了信息。
The description of ‘language model’ is accompanied by an example even a five-year-old can understand.
即使是五岁的孩子也可以理解“语言模型”的示例。
More importantly, Dale’s article includes the properties of GPT-3 that are important to know, in order to carry further exploration. The article mentions the transformer-based architecture of GPT-3, along with the number of parameters within GPT-3 and its predecessor.
更重要的是，Dale的文章包含了一些重要的GPT-3属性，以便进行进一步的探索。 本文提到了基于变压器的GPT-3架构，以及GPT-3及其前身中的参数数量。
The article concludes with statements to indicate the main reason as to why GPT-3 sets itself apart from its counterparts, which is mainly the fact that GPT-3 has a general application to a variety of tasks, without the need for fine-tuning.
本文以声明结尾，指出了GPT-3之所以与众不同的主要原因，这主要是因为GPT-3可以广泛应用于各种任务，而无需进行微调。
本文是以下方面的绝佳读物： (This article is an excellent read for:)
General audiences: The language used in the article is not too technical. The information presented is well structured and presented in a format that enables the understanding of key GPT-3 characteristics and benefits. No major technical background required.
普通读者 ：本文中使用的语言不太技术性。 所提供的信息结构合理，并以能够理解GPT-3关键特征和优势的格式提供。 不需要主要的技术背景。
最新的AI会杀死编码吗？ 弗雷德里克·巴斯勒(Frederik Bussler) (Will The Latest AI Kill Coding? By Frederik Bussler)
Web programmers will want to read this article
Web程序员将要阅读本文
简短评论 (Short Review)
Frederik Bussler has written a piece that has reignited a question that has been around the tech industry for a few years now.
弗雷德里克·巴斯勒(Frederik Bussler )撰写的一篇文章重新引发了一个问题，这个问题已经困扰着科技行业几年。
Frederik presents a brief background into the different versions of GPT and an opinion based reasoning as to why the coding job role might face an AI-Apocolypse sooner than later.
弗雷德里克(Frederik)简要介绍了GPT的不同版本，并提出了基于意见的推理，以解释为什么编码工作角色迟早会遇到AI启示。
详细审查 (Detailed Review)
Frederik starts his latest article with an indication to the reader in regards to how quick the progression of AI advancement can be. He references a question proposed by researchers in 2017, which asks the question that “Could AI write most code by 2040?”. In 2020 we are no doubt beginning to see an answer to that question.
弗雷德里克(Frederik)在他的最新文章中向读者表明了AI进步的速度有多快。 他引用了研究人员在2017年提出的一个问题，该问题是“ 到2040年AI会编写最多的代码吗？”。 毫无疑问，在2020年，我们将开始看到该问题的答案。
Like most other articles describing GPT-3, Frederik mentions the key characteristic of the GPT-3 language model, which is its ability to generalized to domain-specific tasks as a result of its very robust neural network parameters. This, as noted by Frederik, is why GPT-3 can code in practically any programming language.
像大多数其他描述GPT-3的文章一样，Frederik提到了GPT-3语言模型的关键特征，这是由于其非常强大的神经网络参数，它能够将其推广到特定领域的任务。 正如Frederik所指出的，这就是为什么GPT-3可以使用几乎任何编程语言进行编码的原因。
Frederik breaks away from the main question the article is trying to address and provides the reader with a detailed description of the variants of the GPT language model.
Frederik摆脱了本文试图解决的主要问题，并为读者提供了GPT语言模型变体的详细说明。
The article steers back to the main subject, with a section named ‘Evolve or Die’, a faith that most industries will probably be subjected too, but coders more sooner than others.
文章以“进化还是消亡”为主题，回到了主要主题，这种信念认为大多数行业也可能受到影响，但编码人员比其他行业要早得多。
Frederik provides further information to the reader — who might be sceptical at the claims Frederik is making, that ‘No-code’ movement has been in full force with the introduction of platforms such as Squarespace, Wix, GoDaddy, generally WYSIWYG website generating platforms.
弗雷德里克(Frederik)向读者提供了更多信息-弗雷德里克(Frederik)提出的主张可能会持怀疑态度，即“无代码”运动在诸如Squarespace，Wix，GoDaddy之类的平台(通常为所见即所得网站生成平台)的引入下已全面发挥作用。
To back up the trend of the emergence of code-free platforms, Frederik includes AutoML and links to a company that enables the creation of AI models with no coding skills.
为了支持无代码平台出现的趋势，Frederik包括AutoML并链接到一家公司，该公司可以创建没有编码技能的AI模型。
I get the impression that Frederik is pretty convinced that the coding or programmer role is fair game for AI in the very near future.
我的印象是，Frederik非常相信编码或程序员的角色在不久的将来对AI来说是公平的游戏。
Read this opinion-based article and find arguments that could reaffirm your worry of an AI-Apocolypse within several industries.
阅读这篇基于观点的文章，找到可以重申您对多个行业中的AI启示感到担忧的论点。
本文是必读的： (This article is a must-read for:)
Programmers: I’ve always taken comfort in the fact that programmers and coders will be required even though AI and automation will decimate a large number of job roles in the future. The emergence and potential of GPT-3 seem to have ignited some worries that we should probably give more attention to.
程序员 ：尽管AI和自动化将来会淘汰大量工作角色，但仍然需要程序员和编码员，这让我一直感到欣慰。 GPT-3的出现和潜力似乎引起了一些我们可能应该更多关注的担忧。
3分钟解释Ayush Sharma为何对GPT-3进行过度宣传的原因 (A 3 Minute Explainer on Why GPT-3 is Overhyped By Ayush Sharma)
Not everyone is buying what GPT-3 is selling
并非所有人都在购买GPT-3的产品
精简版 (Short Version)
There is no shortage of articles glorifying the feats of GPT-3, but Ayush article seeks to bring us back down to Earth. This bluntly written article is short and convinces the reader that GPT-3 is nothing more than an AI-tool overhyped by mainstream media.
不乏赞美GPT-3壮举的文章，但是Ayush的文章力图使我们重返地球。 这篇简短的文章简短，使读者相信GPT-3仅仅是主流媒体夸大的AI工具。
详细版本 (Detailed Version)
Ayush Sharma article aims to bring individuals in tech that hold the notion that GPT-3 is an advent to the much anticipated Artificial General Intelligence (AGI).
Ayush Sharma的文章旨在使个人认为GPT-3是万众期待的人工智能(AGI)的出现。
I particularly like the analogy he uses where he describes the similarity of the notion embraced by a few individuals in tech, to cavemen delivering they have solved nuclear fusion through the discovery of fire.
我特别喜欢他用比喻来描述一些技术人员所拥护的概念与穴居人的相似之处，因为穴居人通过发现火力解决了核聚变问题。
Ayush seems not to be fooled by the shiny new toys build through GPT-3. The article reminds us that the hype and attention we are currently seeing is a familiar behavioural cycle adopted by mainstream media to overestimate the potential of AI advanced, all to sell stories.
Ayush似乎并没有被通过GPT-3制造的闪亮新玩具所迷住。 这篇文章提醒我们，我们目前看到的炒作和关注是主流媒体采用的熟悉的行为周期，以高估所有高级AI产品的销售潜力。
Describing GPT-3 as a “glorified $10+ auto-complete software” is rather harsh. In this article, Ayush states bluntly reasons as to why GPT-3 is not AGI.
将GPT-3描述为“ 美化了的$ 10 +自动完成软件 ”相当苛刻。 在本文中，Ayush直言不讳地说明了为什么GPT-3不是AGI。
The article goes further to include examples of GPT-3 shortcomings.
本文进一步介绍了GPT-3缺陷的示例。
The conclusion to the article included an embedded tweet from Sam Altman, the CEO of Open AI, where he states that the hype surrounding GPT-3 is not warranted, and although it might be impressive, there is still a long way to go.
文章的结论包括来自Open AI首席执行官Sam Altman的嵌入式推文，他在信中指出，围绕GPT-3的炒作是没有必要的，尽管可能会令人印象深刻，但还有很长的路要走。
本文是以下方面的绝佳读物： (This article is an excellent read for:)
Everyone: It’s easy to get lost in the potential of AI that is showcased across the internet. Therefore, it is always healthy to have a rather critic-like perspective, so as to manage the expectation of the public and prevent future disappointments.
每个人 ：在互联网上展示的AI的潜力很容易迷失。 因此，保持一种颇像批评家的观点总是健康的，以便管理公众的期望并防止未来的失望。
希望您觉得这篇文章有用。 (I hope you found the article useful.)
To connect with me or find more content similar to this article, do the following:
要与我联系或查找更多类似于本文的内容，请执行以下操作：
翻译自: https://towardsdatascience.com/articles-that-will-help-you-understand-gpt-3-610dedc37859
gpt-2 文章自动生成
1165
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
