INDEX
Explanations
key concepts related to media representation and societal issues
New Auto-Interp
Negative Logits
cheid
-0.15
armor
-0.14
anim
-0.14
tract
-0.14
amt
-0.14
onet
-0.14
é»
-0.14
avor
-0.13
ereg
-0.13
ware
-0.13
POSITIVE LOGITS
humble
0.18
spb
0.17
Lands
0.17
idea
0.16
idea
0.15
_pb
0.15
concept
0.15
åħ»
0.15
haf
0.15
ub
0.14
Activations Density 0.207%