INDEX
Explanations
variations of the word "caribou"
New Auto-Interp
Negative Logits
arbeit
-0.16
ught
-0.16
carpet
-0.15
carpets
-0.15
ships
-0.15
arak
-0.14
erdale
-0.14
owski
-0.14
esc
-0.14
aping
-0.14
POSITIVE LOGITS
thers
0.21
ordum
0.20
bone
0.18
bohydr
0.16
Demir
0.16
nage
0.15
cter
0.15
atures
0.15
lsen
0.15
eÅŁ
0.15
Activations Density 0.047%