36 protein LMs: 3 archs x 4 scales x 3 seeds on UniRef50, compute-optimal. Transformers out-scale Mamba SSMs. Code: github.com/ygzdvr/nanoprot
Yagiz Devre
yagizdevre
AI & ML interests
None yet
Organizations
models 15
yagizdevre/nanoprot-mamba-XS
Text Generation • Updated • 7
yagizdevre/nanoprot-mamba-S
Text Generation • Updated • 6
yagizdevre/nanoprot-mamba-M
Text Generation • Updated • 1
yagizdevre/nanoprot-mamba-L
Text Generation • Updated • 3
yagizdevre/nanoprot-gpt2-S
Text Generation • Updated • 3
yagizdevre/nanoprot-gpt2-M
Text Generation • Updated • 1
yagizdevre/nanoprot-gpt2-L
Text Generation • Updated • 13
yagizdevre/nanoprot-esm2-XS
Fill-Mask • Updated • 10
yagizdevre/nanoprot-esm2-S
Fill-Mask • Updated • 2
yagizdevre/nanoprot-esm2-M
Fill-Mask • Updated • 1
datasets 0
None public yet