TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Microsoft and Nvidia have created a 530B parameter language model

7 点作者 axiom92超过 3 年前

1 comment

axiom92超过 3 年前
The model, named Megatron-Turing NLG 530B, is about 3x bigger than GPT-3.<p>The blog post doesn&#x27;t provide a lot of numbers but looks like it beats the state-of-the-art in a couple of commonsense reasoning benchmarks.<p>Still, it shows that you can&#x27;t just keep scaling the models and expect magic.
评论 #28841317 未加载