GPT-4
美国
人工智能GPT-3替代大语模型(LLMS)

GPT-4

下一代GPT模型。目前尚不可用。

标签:
爱站权重:PC 百度权重移动 百度移动权重

OpenAI首席执行官Sam Altman在[AC10在线聚会]的一个问答环节(https://news.knowledia.com/us/en/articles/sam-altman-q--and-a-gpt------- -agi-lesswrong-AA6293DDE7B95B537F5F50E37C861645FD9B4DBB),谈到了即将来临的GPT-4版本。

他在采访中说,与普遍的看法相反,GPT-4不会比GPT-3大,但会使用更多的计算资源。 100万亿个参数模型不会是GPT-4,并且距离遥远。他们从较小的型号中获得了更多的性能。也许他们永远不需要这么大的模型。 [agi](https://en.wikipedia.org/wiki/artcover_general_intelligence)可能需要尚未发生的新算法突破,而不仅仅是建立更大的模型。

GPT-4将更多地关注编码,即[codex](https://gpt3demo.com/apps/openai-codex)(codex是GPT-3的后代)。值得注意的是,OpenAI最近通过私人Beta中的API发布了Codex。 Codex也是[Github Copilot](https://gpt3demo.com/apps/github-copilot)的基础。它理解了十多种语言,还可以用自然语言来解释简单的命令并代表用户执行它们,从而允许在现有应用程序中构建自然语言界面。

进度将来自Openai在GPT的各个方面工作(数据,algos,微调等)。 GPT -4可能能够在更长的上下文中使用,并通过不同的损失功能进行培训 - OpenAI为此具有“视线”。

编码似乎是GPT的另一个主要应用领域。一个这样的例子是Microsoft最近宣布的基于GPT-3的辅助功能,用于公司的[PowerApps软件](https://gpt3demo.com/apps/microsoft-power-fx),将自然语言转换为代码段。随着Altman的最新声明,预计Openai将通过新的GPT进行更多的能力来利用这一能力。

#### GPT-5怎么样?

GPT-5可能能够通过图灵测试。但可能不值得付出努力。

资料来源:
-https://analyticsindiamag.com/gpt-4-sam-altman-confirms-the-rumours/
-https://news.news.knowledia.com/us/en/articles/sam-altman-q-and-a-gpt-and-agi-lesswrong-aa6293dde7b95b95b537f537f537f50e37c861645fd9b4dbb

原文:

Sam Altman, the CEO of OpenAI, in a question-answer session in [AC10 online meetup](https://news.knowledia.com/US/en/articles/sam-altman-q-and-a-gpt-and-agi-lesswrong-aa6293dde7b95b537f5f50e37c861645fd9b4dbb), spoke about the impending GPT-4 release.

He said in the interview that contrary to popular belief, GPT-4 will not be any bigger than GPT-3 but will use more compute resources. 100 trillion parameter model won't be GPT-4 and is far off. They are getting much more performance out of smaller models. Maybe they will never need such a big model. [AGI](https://en.wikipedia.org/wiki/Artificial_general_intelligence) will likely require new algorithmic breakthroughs that haven't happened yet, instead of just building bigger models.

GPT-4 would focus more on coding, that is, [Codex](https://gpt3demo.com/apps/openai-codex) (Codex is a descendant of GPT-3). It is worth noting that OpenAI recently released Codex through API in private beta. Codex is also the basis for [GitHub Copilot](https://gpt3demo.com/apps/github-copilot). It understands more than a dozen languages and can also interpret simple commands in natural language and execute them on users’ behalf, allowing building a natural language interface to existing applications.

The progress will come from OpenAI working on all aspects of GPT (data, algos, fine-tuning, etc.). GPT-4 will likely be able to work with longer context and be trained with a different loss function - OpenAI has "line of sight" for this.

Coding seems to be another major application area for GPTs. One such example is Microsoft’s recently announced GPT-3 based assistive feature for the company’s [PowerApps software](https://gpt3demo.com/apps/microsoft-power-fx) that converts natural language into code snippets. With Altman’s recent statements, it is expected that OpenAI would be leveraging this capability more with the new instalment of GPT.

#### What about GPT-5?

GPT-5 might be able to pass the Turing test. But probably not worth the effort.

Sources:
- https://analyticsindiamag.com/gpt-4-sam-altman-confirms-the-rumours/
- https://news.knowledia.com/US/en/articles/sam-altman-q-and-a-gpt-and-agi-lesswrong-aa6293dde7b95b537f5f50e37c861645fd9b4dbb

数据统计

数据评估

GPT-4浏览人数已经达到328,如你需要查询该站的相关权重信息,可以点击"爱站数据""Chinaz数据"进入;以目前的网站数据参考,建议大家请以爱站数据为准,更多网站价值评估因素如:GPT-4的访问速度、搜索引擎收录以及索引量、用户体验等;当然要评估一个站的价值,最主要还是需要根据您自身的需求以及需要,一些确切的数据则需要找GPT-4的站长进行洽谈提供。如该站的IP、PV、跳出率等!

关于GPT-4特别声明

本站GPT 案例导航提供的GPT-4都来源于网络,不保证外部链接的准确性和完整性,同时,对于该外部链接的指向,不由GPT 案例导航实际控制,在2023年3月9日 下午10:30收录时,该网页上的内容,都属于合规合法,后期网页的内容如出现违规,可以直接联系网站管理员进行删除,GPT 案例导航不承担任何责任。

相关导航