Dwarkesh Podcast:

Analysis: from 2019 to 2025, gains in pretraining compute efficiency came mostly from data improvements rather than model improvements  —  Breaking down 6 years of pretraining progress into data vs model improvements  —  How much of the rapid progress in AI that we've seen over the last few years 1 …

Dwarkesh Podcast:

Analysis: from 2019 to 2025, gains in pretraining compute efficiency came mostly from data improvements rather than model improvements  —  Breaking down 6 years of pretraining progress into data vs model improvements  —  How much of the rapid progress in AI that we've seen over the last few years 1 …

Source: TechMeme
Source Link: https://www.techmeme.com/260908/p49#a260908p49