INDEX
Explanations
images or descriptions of shirts being worn or mentioned
references to shirts, particularly in various contexts
New Auto-Interp
Negative Logits
audi
-0.72
yrinth
-0.69
judicial
-0.68
ntil
-0.66
satellites
-0.65
ths
-0.63
20439
-0.62
ruciating
-0.62
sys
-0.61
ITED
-0.61
POSITIVE LOGITS
shirt
1.16
leeve
1.14
sleeve
1.08
shirts
1.06
shirts
1.05
hirt
1.04
sleeves
1.01
shirt
0.99
cloth
0.96
pins
0.94
Activations Density 0.022%