Skip to content

🎉CUDA 笔记: flash_attn、sgemm、sgemv、warp reduce、block reduce、dot product、elementwise、softmax、layernorm、rmsnorm、hist etc.

active 2024-02-072024-04-14 (UTC)

Partial coverage17,782 / 22,845 hourly files (78%) · 2 absent upstream · 5,061 failed, retryable2024-01-032026-08-12 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
269
Pushes
56
Pull requests
0
Issues
4
Stars
188
Forks
13

Activity over time

Daily event counts in the loaded window

Line chart, 68 days from 2024-02-07 to 2024-04-14. Pushes: 56 total, peak 35 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 4 total, peak 2 in a day. Comments: 4 total, peak 2 in a day. Stars: 188 total, peak 13 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Top contributors

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

ContributorContributionsPushesPRsComments
DefTruth595602
Phoenix82153001
zbt782001

Recent activity

Latest issues, pull requests and releases

  • ReleaseDefTruth2024-03-25 05:57
    v0.3 flash_attn_1_fwd_f32
  • Issue comment#4Phoenix82152024-03-24 07:08
    您好,请问sigmoid算子这里为啥没有考虑指数溢出问题
  • Issue#4Phoenix82152024-03-24 07:08
    您好,请问sigmoid算子这里为啥没有考虑指数溢出问题
  • Issue comment#4DefTruth2024-03-24 01:02
    您好,请问sigmoid算子这里为啥没有考虑指数溢出问题
  • Issue#4Phoenix82152024-03-23 14:06
    您好,请问sigmoid算子这里为啥没有考虑指数溢出问题
  • ReleaseDefTruth2024-03-23 08:22
    CUDA Learn Note v0.2
  • Issue#3DefTruth2024-03-23 03:33
    resources
  • Issue#2zbt782024-03-12 06:41
    layer norm实现
  • Issue comment#2zbt782024-03-12 06:41
    layer norm实现
  • Issue comment#2DefTruth2024-03-12 02:10
    layer norm实现

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 188 stars here means stars gained during the window, not the repo's star count.