List your company

Next-token prediction on OpenWebText (train)

18.4PPLHermite activation
18.357618.643818.9319.2162Feb 3, 2025

Evaluation results

MethodLinks
Hermite activationAct.=Hermite, Deg.=3, FLOPs=87.56G2025.0218.42.9122
Fourier activationAct.=Fourier, Deg.=6, FLOPs=88.69G2025.0218.852.9365
Tropical activationAct.=Tropical, Deg.=6, FLOPs=87.75G2025.0218.952.9418
GELUAct.=GELU, Deg.=null, FLOPs=87.52G2025.0219.052.9471
SiLUAct.=SiLU, Deg.=null, FLOPs=87.37G2025.0219.462.9684