Published May 20, 2024 | Version v0.2.0
Software Open

Accelerated Scan

Description

@unixpickle has fused the sequence reversal required by backward into the kernel and vectorized loads and stores to load entries, training is 30-40 percent faster on 3090.

Notes

If you use Accelerated Scan, please cite it as below.

Files

proger/accelerated-scan-v0.2.0.zip

Files (78.6 kB)

Name Size Download all
md5:b903cd8691db6d7f61f9e02b949eaf64
78.6 kB Preview Download

Additional details

Related works