INDEX
Explanations
names often followed by surnames
New Auto-Interp
Negative Logits
ial
0.75
ogen
0.72
let
0.71
attr
0.70
olog
0.69
therapy
0.69
oring
0.68
ilt
0.66
group
0.66
тельной
0.65
POSITIVE LOGITS
Messi
1.13
Messi
1.12
Beyoncé
1.05
Godzilla
1.05
Skywalker
1.04
Pikachu
1.02
Picasso
1.01
Bezos
1.00
Schwarzenegger
0.99
Putin
0.99
Activations Density 0.474%