INDEX
Explanations
proper nouns or names with the string "app"
occurrences of the substring "app"
New Auto-Interp
Negative Logits
fright
-0.65
ging
-0.58
Unreal
-0.58
deaf
-0.58
multim
-0.58
SOURCE
-0.56
blind
-0.56
vae
-0.56
compan
-0.55
psychiat
-0.55
POSITIVE LOGITS
enegger
1.07
erella
1.06
enburg
1.05
enhagen
1.04
earance
1.04
enger
1.04
erer
1.00
enberg
0.99
etts
0.99
erton
0.96
Activations Density 0.045%