Heykuki News

TopNewBestAskShowJobs
TopNewBestAskShowJobs
1.
Power Attention: Efficient CUDA Kernels for Symmetric Power Transformers (github.com/m-a-n-i-f-e-s-t)
6 points
txus
a year ago
2 comments
2.
PowerRetention: a drop-in replacement for FlashAttention in LLMs (github.com/m-a-n-i-f-e-s-t)
2 points
dvrp
8 months ago
2 comments