BLaST: High Performance Inference and Pretraining using BLock Sparse Transformers
Paper
•
2507.03117
•
Published
Sameer Deshmukh
v0dro
·
AI & ML interests
None yet
Recent Activity
new activity
22 days ago
open-thoughts/OpenThoughts3-1.2M:why is it all code data? where's math and science
authored
a paper
5 months ago
BLaST: High Performance Inference and Pretraining using BLock Sparse
Transformers
Organizations
why is it all code data? where's math and science
1
#7 opened about 1 month ago
by
Gavin-Wang
authored
a
paper
5 months ago