INDEX
Explanations
the word "so" used as a conjunction or intensifier
New Auto-Interp
Negative Logits
ston
-0.16
ead
-0.15
.statusText
-0.15
wald
-0.15
roman
-0.15
awan
-0.15
lica
-0.14
ton
-0.14
ster
-0.14
regnum
-0.14
POSITIVE LOGITS
-called
0.19
zelf
0.18
stride
0.17
chy
0.16
unes
0.15
yen
0.14
884
0.14
yntax
0.14
iner
0.14
opic
0.14
Activations Density 0.034%