INDEX
Explanations
references to bars, nightlife, and social venues
New Auto-Interp
Negative Logits
ValueStyle
-0.48
مشين
-0.40
ANDUM
-0.40
بالن
-0.37
vectoriales
-0.35
motyw
-0.35
AssemblyCompany
-0.35
torchvision
-0.35
Спољашње
-0.34
stdafx
-0.34
POSITIVE LOGITS
tavern
0.78
bartender
0.77
pubs
0.75
beer
0.73
pub
0.73
drinks
0.67
Tavern
0.65
booze
0.65
drink
0.65
bart
0.65
Activations Density 0.015%