INDEX
Explanations
mentions of clothing items, particularly shirts
references to shirts and clothing-related terms
New Auto-Interp
Negative Logits
ingred
-0.67
sys
-0.64
ths
-0.64
fert
-0.64
prey
-0.63
SPONSORED
-0.63
satellites
-0.62
subsequ
-0.60
elsen
-0.59
nesota
-0.59
POSITIVE LOGITS
sleeve
1.26
shirt
1.21
sleeves
1.21
leeve
1.17
hirt
1.17
shirts
1.13
shirt
1.10
shirts
1.04
Shirt
1.03
collar
1.02
Activations Density 0.046%