INDEX
Explanations
references to the name "Smith"
New Auto-Interp
Negative Logits
ipar
-0.16
coni
-0.15
uchar
-0.15
apia
-0.14
aison
-0.14
nings
-0.14
erras
-0.14
دÙĩ
-0.14
oga
-0.14
pag
-0.14
POSITIVE LOGITS
sonian
0.34
son
0.25
ies
0.22
urst
0.22
wick
0.21
ers
0.21
sons
0.19
boro
0.18
urai
0.18
s
0.17
Activations Density 0.010%