INDEX
Explanations
mentions of locations, particularly the state of Arizona
repeated instances of the name "Ari" and related proper nouns
New Auto-Interp
Negative Logits
ruary
-0.73
acebook
-0.72
nesday
-0.69
ishers
-0.68
leneck
-0.68
omsky
-0.67
eele
-0.67
shave
-0.66
Toro
-0.63
rers
-0.62
POSITIVE LOGITS
Pacific
0.81
hya
0.78
gov
0.76
Sinai
0.75
Khan
0.74
etus
0.71
Wan
0.71
ī
0.69
eta
0.68
Bir
0.68
Activations Density 0.096%