INDEX
Explanations
descriptions related to how someone is dressed
references to clothing or how individuals are attired
New Auto-Interp
Negative Logits
nir
-0.75
SHIP
-0.74
orio
-0.68
Focus
-0.66
ium
-0.63
":["
-0.61
vd
-0.61
Ascend
-0.60
platform
-0.59
existence
-0.59
POSITIVE LOGITS
rehearsal
1.01
gown
0.95
rehears
0.94
impecc
0.88
robes
0.86
casually
0.84
appropriately
0.82
uniforms
0.82
differently
0.82
ident
0.81
Activations Density 0.065%