Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Tencent Hunyuan-Large.
They use
- 16 experts, of which one is activated per token
- 1 shared expert that is always active
in summary that makes around 52B active parameters per token instead of the 405B of LLama3.1.
There are no comments on this comment yet.
They use
- 16 experts, of which one is activated per token
- 1 shared expert that is always active
in summary that makes around 52B active parameters per token instead of the 405B of LLama3.1.