INDEX
Explanations
mentions of the word 'skinny', particularly in the context of physical appearance or body size
repeated references to "skinny."
New Auto-Interp
Negative Logits
aga
-0.83
uable
-0.79
ians
-0.78
iations
-0.74
itaire
-0.72
ente
-0.72
amaz
-0.69
emin
-0.69
itated
-0.69
arding
-0.66
POSITIVE LOGITS
OA
0.91
ETF
0.81
xual
0.79
ĨĴ
0.75
ODE
0.73
ño
0.73
JA
0.70
eki
0.70
ESSION
0.70
Filename
0.69
Activations Density 0.036%