News for hackers. Stuff that matters.

A comment on The Era of 1-bit LLMs: ternary parameters for cost-effective computing. Parent comment

Re:The Era of 1-bit LLMs: ternary parameters for cost-effect... by Noe2097 on Wednesday February 28, 2024 @05:17PM (#39544181)
There is another _shocking_ realization in this work: there are 11 types of people: those who know what binary means, those who don't, and those who say they do but actually don't.

"The era of 1-bit LLMs"

Representing { -1, 0, 1 } can't be done with 1-bit, I'm sorry -- and sad, please let's all get back to something vaguely sound and rigorous.

[ Parent | on HN ]
Re:The Era of 1-bit LLMs: ternary parameters for cost-effect... by npunt on Wednesday February 28, 2024 @05:20PM (#39544223)
Ternary supporters are always bitter about this

(I'll let myself out)

[ Parent | on HN ]
Re:The Era of 1-bit LLMs: ternary parameters for cost-effect... by esrauch on Wednesday February 28, 2024 @08:52PM (#39545745)
One trit but that's not a word anyone knows.
[ Parent | on HN ]
Re:The Era of 1-bit LLMs: ternary parameters for cost-effect... by baq on Thursday February 29, 2024 @01:17AM (#39546827)
That used to be true yesterday…
[ Parent | on HN ]
Re:The Era of 1-bit LLMs: ternary parameters for cost-effect... by gpderetta on Thursday February 29, 2024 @11:05AM (#39551173)
There are 10 types of people, those who don't know binary, those who do and those who know ternary.
[ Parent | on HN ]
Re:The Era of 1-bit LLMs: ternary parameters for cost-effect... by hk__2 on Thursday February 29, 2024 @12:07PM (#39552105)
> please let's all get back to something vaguely sound and rigorous

Something rigorous would be to actually read the paper rather than stop at the first part of its title. The authors are not claiming their LLM is 1-bit.

[ Parent | on HN ]