INDEX
Explanations
proper nouns related to sports teams and athletes
entities or terms related to names and titles
New Auto-Interp
Negative Logits
olitan
-0.83
ynthesis
-0.75
creen
-0.69
Coliseum
-0.67
aukee
-0.67
advisor
-0.64
ARD
-0.62
ourt
-0.61
hips
-0.60
aii
-0.59
POSITIVE LOGITS
gaard
1.08
theless
0.95
xual
0.84
heid
0.81
hurst
0.80
eering
0.78
thren
0.75
bol
0.75
ĸļ
0.75
eem
0.74
Activations Density 0.036%