INDEX
Explanations
references to sports jerseys and uniforms
New Auto-Interp
Negative Logits
enthal
-0.83
moderate
-0.79
sth
-0.78
course
-0.77
audi
-0.75
lishes
-0.75
yrinth
-0.75
llor
-0.74
hower
-0.74
ebin
-0.73
POSITIVE LOGITS
worn
1.21
jerseys
1.16
jersey
1.08
sleeves
1.06
sleeve
1.04
seys
0.97
adorned
0.92
collar
0.90
tint
0.87
wearer
0.85
Activations Density 0.012%