arxiv.org favicon

[2302.13971] LLaMA: Open and Efficient Foundation Language Models

5
公开标注数
5
参与人数
2026-07-21 09:17:07
首次 Whisper

本页的公开 Whisper

划选高亮2026-07-21 12:38:07
原文高亮摘录
possible to train state-of-the-art models using p
Whisper 随想笔记
wait, 13B beating GPT-3? that's... suspiciously good
划选高亮2026-07-21 12:29:07
原文高亮摘录
possible to train state-of-the-art models using p
Whisper 随想笔记
finally, open models that don't need those secret datasets
划选高亮2026-07-21 09:35:07
原文高亮摘录
a collection of foundation language models ranging from 7B to 65B parameters
Whisper 随想笔记
Trillions of tokens on public data? I'm skeptical but curious.
划选高亮2026-07-21 09:26:07
原文高亮摘录
a collection of foundation language models ranging from 7B to 65B parameters
Whisper 随想笔记
Finally open models that beat GPT-3, that's wild.
划选高亮2026-07-21 09:17:07
原文高亮摘录
a collection of foundation language models ranging from 7B to 65B parameters
Whisper 随想笔记
7B to 65B is a nice range, but 65B still needs a lot of GPUs.

分享本页 Whisper

分享到 X
短链接
https://domwhisper.com/s/3972039dbf3f
嵌入代码
<iframe src="https://domwhisper.com/embed/3972039dbf3f" width="100%" height="480" style="border:0;border-radius:16px" loading="lazy"></iframe>

看看大家在 arxiv.org 上讨论了什么

安装 DomWhisper,浏览网页时实时查看 whisper,也可以加入讨论。

获取插件