INDEX
Explanations
phrases related to the fictional village of Sweetaven
references to the word "Swe" and its variations
New Auto-Interp
Negative Logits
PowerPoint
-0.82
LIFE
-0.73
ional
-0.70
gravity
-0.68
ocl
-0.68
geography
-0.66
misinformation
-0.65
otherwise
-0.64
Lith
-0.64
LIB
-0.64
POSITIVE LOGITS
arer
1.22
swe
1.12
izzard
1.05
eping
1.02
Swe
0.99
arers
0.98
zza
0.97
atered
0.97
itzer
0.96
onder
0.94
Activations Density 0.010%