Activity
Stars and mentions · Last 30 days, Oct 6 – Nov 4
Top repositories
marin-community’s repositories ranked by pull request authors, as on trending.
Contributes to
Repositories outside marin-community’s account where marin-community opened or reviewed pull requests in the past 365 days.
Who noticed marin-community
Everyone who starred or forked marin-community’s repositories since the archive began in 2011.
Notable accounts
Ordered by the stars their own repositories have received.
Biggest fans
Accounts that starred the most of these repositories.
Breakout weeks
Each repository’s best week of new stars, with the release published around it.
Achievements
Awards earned by the repositories marin-community owns. Select an award to embed its badge.
No awards yet
Repositories earn awards by reaching the top three in a completed trending day or week.
A repository earns an award for finishing first, second or third in a completed trending day or week, ranked by new stars, new forks or pull request authors. Awards follow a repository through renames and transfers and count toward its current owner. Weeks run Monday to Sunday, UTC.
Mentions
@Sxandchenxing @micahgoldblum The Marin folks tried Hyperball and still found z-loss crucial in preventing loss from blowing up! marin#8227 @classiclarryd
I was really impressed by Marin's published evals policies: marin#9409 Precise and very clearly written.
@CatherineeLii cool experiment! i was wondering whether you believe this high-data-budget fine-tuning gap will persist across batch sizes + weight decays? many researchers find that Muon's critical batch size is much higher alphaxiv.org marin#1565 cc @leloykun t.co
3/3 Finding the right hinting primitives for agentic benchmarks (longer, messier rollouts) is an open question. There’s already some work happening on this front in Marin! Issue: marin#9541
Of course this is retrospective. Hindsight etc. etc. That said, it was basically "do what we did for the Paloma fit". No 535b losses need have been consulted. As always, code and context are on GitHub marin-logbooks#8
We’re Marin, the details are pretty much always available! marin#9960 This time we’re deploying Triton router top-k and short conv, leaner ragged a2a transport, fused expert backward, and XLA scheduling fixes