Analysis: from 2019 to 2025, gains in pretraining compute efficiency came mostly from data improvements rather than model improvements (Dwarkesh Podcast)
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 …
We haven't written up this one. Techmeme has the full story — the link below goes straight to it.