INDEX
Explanations
elements related to superhero narratives and their interactions
New Auto-Interp
Negative Logits
osi
-0.15
pac
-0.14
istrovstvÃŃ
-0.14
ãĥ³ãĤº
-0.14
dbus
-0.14
FUN
-0.13
çĭ¬ç«ĭ
-0.13
алеж
-0.13
ØŃÙĬ
-0.13
orm
-0.13
POSITIVE LOGITS
pil
0.16
exchange
0.15
otope
0.14
exchange
0.14
fight
0.14
suic
0.14
è£ķ
0.14
clim
0.14
Descriptors
0.14
teleport
0.14
Activations Density 0.051%