Skip to content

maximal update parametrization (µP) from https://github.com/microsoft/mup

active 2023-08-152026-06-03 (UTC)

Complete coverage26,394 / 26,394 hourly files (100%) · 2 absent upstream2023-08-152026-08-18 (UTC)
Events
777
Pushes
4
Pull requests
11
Issues
24
Stars
662
Forks
38

Activity over time

Daily event counts in the loaded window

Line chart, 1024 days from 2023-08-15 to 2026-06-03. Pushes: 4 total, peak 2 in a day. Pull requests: 11 total, peak 2 in a day. Issues: 24 total, peak 2 in a day. Comments: 36 total, peak 5 in a day. Stars: 662 total, peak 33 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

Recent activity

Latest issues, pull requests and releases

  • Pull request#81francois-rozet2025-09-28 17:21
  • Pull request#82jmerkow2025-03-24 17:11
  • Issue#78josephcappadona2024-12-24 20:53
    How to use with SSL methods like DINOv2?
  • Issue comment#81francois-rozet2024-10-18 21:53
    Add support for dataclasses
  • Issue comment#81microsoft-github-policy-service[bot]2024-10-18 21:51
    Add support for dataclasses
  • Issue comment#81microsoft-github-policy-service[bot]2024-10-18 21:51
    Add support for dataclasses
  • Pull request#81francois-rozet2024-10-18 21:51
  • Issue comment#1karibbov2024-10-04 16:32
    Consider decoupled weight decay optimizers?
  • Pull request#79JeremyCCHsu2024-09-05 04:59
  • Issue comment#79microsoft-github-policy-service[bot]2024-09-05 04:58
    CNN utility
  • Pull request#79JeremyCCHsu2024-09-05 04:58
  • Issue#78josephcappadona2024-07-30 17:50
    How to use with SSL methods like DINOv2?
  • Issue#77norikazu992024-07-26 02:43
    MuP for RNNs
  • Issue#76gordicaleksa2024-07-19 15:27
    Not getting perf improvements from muP at ~1.5B scale
  • Pull request#75emergenz2024-07-17 11:54
  • Issue comment#1SeunghyunSEO2024-07-11 05:15
    Consider decoupled weight decay optimizers?
  • Issue comment#71ofivite2024-06-21 11:28
    Increasing coord check for the network output
  • Issue comment#71SeunghyunSEO2024-06-21 06:56
    Increasing coord check for the network output
  • Issue#74norikazu992024-05-30 08:24
    MuP for Mamba
  • Issue comment#73thomasfortin12024-05-06 02:50
    Refactor: Addressing Sources of User Error
  • Issue comment#73microsoft-github-policy-service[bot]2024-05-06 02:48
    Refactor: Addressing Sources of User Error
  • Pull request#73thomasfortin12024-05-06 02:48
  • Issue comment#72janEbert2024-05-03 14:35
    Support FSDP usage
  • Issue comment#72microsoft-github-policy-service[bot]2024-05-03 14:07
    Support FSDP usage
  • Pull request#72janEbert2024-05-03 14:07

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 662 stars here means stars gained during the window, not the repo's star count.