Skip to content

Scaling Data-Constrained Language Models

active 2023-08-172026-02-28 (UTC)

Complete coverage26,338 / 26,338 hourly files (100%) · 2 absent upstream2023-08-152026-08-16 (UTC)
Events
241
Pushes
9
Pull requests
0
Issues
10
Stars
199
Forks
7

Activity over time

Daily event counts in the loaded window

Line chart, 927 days from 2023-08-17 to 2026-02-28. Pushes: 9 total, peak 2 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 10 total, peak 2 in a day. Comments: 13 total, peak 3 in a day. Stars: 199 total, peak 15 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

ContributorContributionsPushesPRsComments
Muennighoff16907
borgr6002
SeunghyunSEO3002
wmn-2313143001
orena12001
thu-yao-01-luo2000

Recent activity

Latest issues, pull requests and releases

  • Issue comment#15wmn-2313142025-06-30 02:14
    Possible incorrect R**2 computation in parametric_fit
  • Issue#15wmn-2313142025-06-30 02:14
    Possible incorrect R**2 computation in parametric_fit
  • Issue comment#15Muennighoff2025-06-27 23:46
    Possible incorrect R**2 computation in parametric_fit
  • Issue#15wmn-2313142025-06-27 22:42
    Possible incorrect R**2 computation in parametric_fit
  • Issue comment#14Muennighoff2025-02-20 07:14
    Validation loss vs model size per step
  • Issue comment#14orena12025-02-20 06:56
    Validation loss vs model size per step
  • Issue comment#14Muennighoff2025-02-20 06:01
    Validation loss vs model size per step
  • Issue#14orena12025-02-20 03:44
    Validation loss vs model size per step
  • Issue comment#13SeunghyunSEO2024-07-23 09:24
    wonder if LR=1e-3 for mup is optimal value from small-scale proxy model and dropout is crucial for multi-epoch
  • Issue comment#13Muennighoff2024-07-22 19:26
    wonder if LR=1e-3 for mup is optimal value from small-scale proxy model and dropout is crucial for multi-epoch
  • Issue comment#13SeunghyunSEO2024-07-22 17:05
    wonder if LR=1e-3 for mup is optimal value from small-scale proxy model and dropout is crucial for multi-epoch
  • Issue comment#13Muennighoff2024-07-22 15:49
    wonder if LR=1e-3 for mup is optimal value from small-scale proxy model and dropout is crucial for multi-epoch
  • Issue#13SeunghyunSEO2024-07-22 08:26
    wonder if LR=1e-3 for mup is optimal value from small-scale proxy model and dropout is crucial for multi-epoch
  • Issue#12thu-yao-01-luo2024-03-23 02:31
    Question about the smoothing method applied over Figure 4.
  • Issue#12thu-yao-01-luo2024-03-22 03:08
    Question about the smoothing method applied over Figure 4.
  • Issue comment#11borgr2023-11-12 01:54
    Flop contour
  • Issue#11borgr2023-11-12 01:54
    Flop contour
  • Issue comment#11Muennighoff2023-11-10 22:52
    Flop contour
  • Issue#11borgr2023-11-10 22:50
    Flop contour
  • Issue#10borgr2023-11-10 22:13
    Names was never defined
  • Issue comment#10borgr2023-11-10 22:12
    Names was never defined
  • Issue comment#10Muennighoff2023-11-09 23:35
    Names was never defined
  • Issue#10borgr2023-11-09 22:08
    Names was never defined

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 199 stars here means stars gained during the window, not the repo's star count.