Mudith Jayasekara(@mudithj):This is how we support a bunch of neolabs training frontier models (some of which we should be able to talk about in public soon when they release their models). Pretty neat trick @part_harry_ and @stefanopopoulos cooked up reading the compressed block directly rather than scanning every byte.

Mudith Jayasekara

@mudithj

co-founded @parsedlabs before we started hill sprinting @baselabs @baseten, half eng/cs phd @rhodes_trust @UniofOxford

Joined January 2018

404 Following 690 Followers

Mudith Jayasekara@mudithj

2026.09.08 21:35

This is how we support a bunch of neolabs training frontier models (some of which we should be able to talk about in public soon when they release their models). Pretty neat trick @part_harry_ and @stefanopopoulos cooked up reading the compressed block directly rather than scanning every byte.