INDEX
Explanations
the term "White" in various contexts
New Auto-Interp
Negative Logits
alm
-0.18
niej
-0.17
uyen
-0.16
epad
-0.16
rian
-0.16
istic
-0.15
огод
-0.15
ừa
-0.15
blackColor
-0.15
vej
-0.15
POSITIVE LOGITS
-collar
0.20
prints
0.20
-white
0.19
fish
0.18
papers
0.18
supremacist
0.18
hall
0.18
aker
0.17
board
0.17
acre
0.17
Activations Density 0.037%