INDEX
Explanations
phrases related to someone being out or away from a certain place or activity
instances of the word "out."
New Auto-Interp
Negative Logits
Syd
-0.64
antine
-0.64
transitions
-0.63
phrine
-0.62
hedral
-0.61
etary
-0.60
Dare
-0.58
gallery
-0.55
iosity
-0.55
kefeller
-0.54
POSITIVE LOGITS
stretched
1.47
fitted
1.32
doing
1.06
fitting
1.05
raged
1.04
done
0.96
bur
0.96
smart
0.96
doors
0.92
ranged
0.92
Activations Density 0.048%