HackerTrans
TopNewTrendsCommentsPastAskShowJobs

rawsh

47 karmajoined 6 years ago

Submissions

Flash-MSA: Accelerating Million-Token Training with Sparse Attention Kernels

nanduruganesh.github.io
34 points·by rawsh·11 hours ago·3 comments

Zyphra ZAYA1-base: First large-scale model trained on AMD

zyphra.com
6 points·by rawsh·8 months ago·1 comments

Debugging divergence between engine and transformers logprobs for RL

gist.github.com
2 points·by rawsh·10 months ago·0 comments