INDEX
Explanations
mentions of the character "Minnie Mouse"
instances of the word "Min" and its variations
New Auto-Interp
Negative Logits
enhanced
-0.62
dove
-0.61
Compass
-0.60
almonds
-0.60
links
-0.59
red
-0.59
migrating
-0.59
upd
-0.59
fast
-0.58
ding
-0.58
POSITIVE LOGITS
istries
1.49
erva
1.22
iatures
1.22
amoto
1.13
imal
1.12
assian
1.08
neapolis
1.07
imize
1.05
oan
1.02
usc
0.98
Activations Density 0.010%