INDEX
Explanations
references to the word "ball" followed by a specific word
mentions of the name "Ball."
New Auto-Interp
Negative Logits
profit
-0.75
iosyncr
-0.69
liness
-0.68
éĢ
-0.65
ktop
-0.64
ä½ľ
-0.64
éĹĺ
-0.63
ths
-0.62
åº
-0.62
ctuary
-0.60
POSITIVE LOGITS
oons
1.10
park
0.96
oon
0.91
haus
0.85
aign
0.85
ball
0.85
guns
0.80
fish
0.79
antine
0.79
istically
0.79
Activations Density 0.010%