INDEX
Explanations
references to the website "Kotaku" and potentially similar terms
repeated mentions of the names "Kot" and "Kap"
New Auto-Interp
Negative Logits
xual
-0.80
IBLE
-0.77
tremend
-0.76
phrine
-0.73
iments
-0.71
uration
-0.69
omething
-0.67
mosqu
-0.67
afort
-0.67
mingham
-0.66
POSITIVE LOGITS
lar
0.95
aepernick
0.83
kat
0.81
Rowling
0.79
zig
0.78
Osw
0.77
zeb
0.72
pps
0.72
emp
0.71
Noon
0.68
Activations Density 0.037%