Skip to content

active 2026-04-232026-07-03 (UTC)

Complete coverage26,681 / 26,681 hourly files (100%) · 2 absent upstream2023-08-152026-08-30 (UTC)
Events
81
Pushes
29
Pull requests
10
Issues
11
Stars
0
Forks
0

Activity over time

Daily event counts in the loaded window

Line chart, 72 days from 2026-04-23 to 2026-07-03. Pushes: 29 total, peak 10 in a day. Pull requests: 10 total, peak 5 in a day. Issues: 11 total, peak 7 in a day. Comments: 8 total, peak 8 in a day. Stars: 0 total, peak 0 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
AlliedToasters5929107
jshaofa-ui1001

Recent activity

Latest issues, pull requests and releases

  • Pull request#83AlliedToasters2026-06-14 22:41
  • Pull request#76AlliedToasters2026-04-26 21:53
  • Issue comment#36jshaofa-ui2026-04-26 18:27
    Preflight probe runs cold — add warmup pass before timing
  • Pull request#71AlliedToasters2026-04-26 04:27
  • Issue#68AlliedToasters2026-04-25 23:41
    Namespace buffer_path by sweep ID to isolate sequential sweeps
  • Pull request#66AlliedToasters2026-04-25 20:29
  • Issue#65AlliedToasters2026-04-25 20:15
    Allow callbacks to emit ragged (variable-length per-sample) tensors via Emit.sample_lengths
  • Pull request#64AlliedToasters2026-04-25 01:32
  • Issue#59AlliedToasters2026-04-24 22:55
    Per-microbatch progress on last-in-chunk layer: writes can stall many minutes between tqdm updates
  • Pull request#60AlliedToasters2026-04-24 22:55
  • Issue#59AlliedToasters2026-04-24 22:34
    Per-microbatch progress on last-in-chunk layer: writes can stall many minutes between tqdm updates
  • Issue#56AlliedToasters2026-04-24 20:23
    Untimed post-loop drain/unload gap between last tqdm update and teardown events
  • Issue#57AlliedToasters2026-04-24 20:23
    Timing coverage gaps: full audit of untimed regions in Sweep.run()
  • Pull request#58AlliedToasters2026-04-24 20:17
  • Issue#48AlliedToasters2026-04-24 17:30
    Lazy forward: exit sweep once all requested captures are emitted
  • Pull request#41AlliedToasters2026-04-24 04:34
  • Issue#32AlliedToasters2026-04-24 03:27
    Partial GPU weight residency for models that almost fit in VRAM
  • Pull request#35AlliedToasters2026-04-24 02:43
  • Pull request#35AlliedToasters2026-04-24 02:11
  • Issue#34AlliedToasters2026-04-24 00:23
    You should have chunks with more than one layer
  • Issue#10AlliedToasters2026-04-23 22:24
    Variable-length input batches: relax fixed seq_len assumption
  • Issue#19AlliedToasters2026-04-23 17:32
    harness_adapter: return contract drifts from benchmark_offload_backends.py (fingerprint shape, VRAM units)

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 0 stars here means stars gained during the window, not the repo's star count.