INDEX
Explanations
words related to running marathons
mentions of marathons and related events
New Auto-Interp
Negative Logits
orum
-0.77
metics
-0.75
oshop
-0.73
imated
-0.70
Vert
-0.69
suscept
-0.68
spr
-0.66
under
-0.65
gypt
-0.65
otle
-0.65
POSITIVE LOGITS
runner
1.03
Runner
0.95
marathon
0.93
runners
0.91
Marathon
0.82
athon
0.81
Polo
0.79
bombing
0.77
NING
0.77
bombings
0.72
Activations Density 0.014%