shikhar
@encapsulated007
Followers
1K
Following
70K
Media
420
Statuses
6K
squeezing FLOPS @lossfunk
0xCAFFEINE
Joined November 2020
just not today, Mr. Hotz!!
Cool competition! `examples/torch_cuda_kernel.py` shows how to use tinygrad (with BEAM=2). The bitter lesson always wins in the end, any tricks people find should be added to our search.
38
12
670
RT @rudythetechy: # Making My Own Girl Gang. Main purpose is to .- share daily progress with each other ( what did u do today ) .- help pro….
0
2
0
RT @anindyadeeps: @teortaxesTex Not really, i have seen these guys writing faster kernels themselves first and getting really amazing speed….
0
1
0
RT @Optimus_Prymm: What's cooking 🔪🍳 @encapsulated007 . Whatever it is just don't be cooked. .
0
1
0
0
2
0
RT @elliotarledge: "mom how did we get so rich".instead of posting cuda slop all day, dad turned up the music, stopped relying on caffiene,….
0
2
0
0
1
0
RT @leloykun: Fast, Numerically Stable, and Auto-Differentiable Spectral Clipping via Newton-Schulz Iteration. Hi all, I'm bacc. I have a l….
0
34
0
RT @finbarrtimbers: I wrote a simple (approximate) MFU calculator, as I find myself doing this calculation often but couldn't find another….
0
8
0
RT @m_sirovatka: @encapsulated007 @salykova_ @cHHillee @vitransformer @lossfunk @paraschopra What Horace is implying is that your kernel is….
0
1
0
RT @navbenny: This is one of the most exciting works being done at @lossfunk by @encapsulated007, @vitransformer. Custom Cuda kernels tail….
0
1
0