INDEX
Explanations
references to the Air Force
references to the Air Force
New Auto-Interp
Negative Logits
apest
-0.84
taboola
-0.78
Seller
-0.75
selves
-0.74
spir
-0.72
ç«
-0.70
izers
-0.67
axter
-0.67
flush
-0.66
theless
-0.66
POSITIVE LOGITS
Academy
1.03
Reserve
1.00
Awakens
0.93
colonel
0.93
Corps
0.92
uniforms
0.82
generals
0.82
Base
0.81
brass
0.80
recru
0.80
Activations Density 0.026%