Skip to content

Accelerated First Order Parallel Associative Scan

active 2024-01-102026-01-10 (UTC)

Complete coverage26,645 / 26,645 hourly files (100%) · 2 absent upstream2023-08-152026-08-29 (UTC)
Events
296
Pushes
27
Pull requests
7
Issues
13
Stars
186
Forks
8

Activity over time

Daily event counts in the loaded window

Line chart, 732 days from 2024-01-10 to 2026-01-10. Pushes: 27 total, peak 6 in a day. Pull requests: 7 total, peak 1 in a day. Issues: 13 total, peak 1 in a day. Comments: 35 total, peak 5 in a day. Stars: 186 total, peak 29 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
proger5327417
unixpickle8025
jeromeku4003
srush3003
WeihanLikk3002
xaviergonzalez3001
kklemon2001
Jokeren2002
ekellbuch2010
neojp1000
ruoyxue1000
JohnAlphaIII1000
kongkiatchanu1001

Recent activity

Latest issues, pull requests and releases

  • Pull request#15ekellbuch2026-01-07 07:19
  • Issue comment#15proger2026-01-07 07:19
    complex: fix inter-block carry in backward_scan and expand test coverage
  • Releaseproger2025-12-25 04:21
    0.3.0 — accelerated_scan.complex
  • Issue#10proger2025-12-25 04:11
    Question: integration with complex numbers
  • Issue#13xaviergonzalez2025-01-21 21:18
    Sequence lengths other than powers of 2?
  • Issue comment#13xaviergonzalez2025-01-21 21:18
    Sequence lengths other than powers of 2?
  • Issue comment#14WeihanLikk2024-12-22 02:41
    gates (A matrix) with a shape of batch * dim * dim * seqlen
  • Issue comment#14proger2024-12-22 02:39
    gates (A matrix) with a shape of batch * dim * dim * seqlen
  • Issue comment#14WeihanLikk2024-12-18 03:36
    gates (A matrix) with a shape of batch * dim * dim * seqlen
  • Issue comment#14proger2024-12-18 03:34
    gates (A matrix) with a shape of batch * dim * dim * seqlen
  • Issue#14WeihanLikk2024-12-10 05:06
    gates (A matrix) with a shape of batch * dim * dim * seqlen
  • Issue comment#13proger2024-11-27 06:17
    Sequence lengths other than powers of 2?
  • Issue#13xaviergonzalez2024-11-23 02:08
    Sequence lengths other than powers of 2?
  • Issue#12JohnAlphaIII2024-10-22 11:15
    Crash with fp16 and bf16
  • Issue comment#5kklemon2024-10-10 20:25
    Is there any solution to support variable sequence length during training?
  • Issue comment#11proger2024-10-05 17:07
    Log-space version
  • Issue#11kklemon2024-10-04 14:45
    Log-space version
  • Issue comment#10proger2024-09-21 23:23
    Question: integration with complex numbers
  • Issue#10ekellbuch2024-09-11 17:19
    Question: integration with complex numbers
  • Pull request#9proger2024-08-20 05:17
  • Releaseproger2024-05-20 10:34
    0.2.0 — faster training!
  • Pull request#7proger2024-03-21 09:25
  • Issue comment#7proger2024-03-21 09:25
    Fused backward pass kernel
  • Issue comment#7unixpickle2024-03-21 02:31
    Fused backward pass kernel
  • Issue comment#7proger2024-03-20 19:40
    Fused backward pass kernel

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 186 stars here means stars gained during the window, not the repo's star count.