Wouldn't seeding the RNG used to pick the next token be more configurable? How w... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		extraduder_ire 7 months ago \| parent \| context \| favorite \| on: Python lib generates its code on-the-fly based on ... Wouldn't seeding the RNG used to pick the next token be more configurable? How would changing the hardware/other software make a difference to what comes out of the model?

tibbar 7 months ago [–]

> Wouldn't seeding the RNG used to pick the next token be more configurable?

Sure, that would work.

> How would changing the hardware/other software make a difference to what comes out of the model?

Floating point arithmetic is not entirely consistent between different GPUs/TPUs/operating systems.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact