INDEX
Explanations
phrases related to a specific political slogan ("Make America Great Again")
commands and calls to action
New Auto-Interp
Negative Logits
kered
-0.73
roman
-0.70
uits
-0.67
nesday
-0.66
cffffcc
-0.66
dk
-0.64
rone
-0.63
*/(
-0.62
mith
-0.62
utton
-0.62
POSITIVE LOGITS
Yourself
1.15
ings
1.10
Your
0.93
ments
0.90
yourselves
0.89
ingly
0.87
away
0.79
ables
0.79
yon
0.77
Vector
0.75
Activations Density 0.165%