INDEX
Explanations
references to military operations and related activities in Afghanistan
New Auto-Interp
Negative Logits
TRL
-0.17
babe
-0.15
ettes
-0.14
èĬ³
-0.14
anja
-0.14
tö
-0.14
shade
-0.14
amen
-0.14
reat
-0.14
zac
-0.14
POSITIVE LOGITS
ason
0.17
oki
0.16
ivo
0.15
.psi
0.14
nemonic
0.14
opt
0.14
728
0.14
нед
0.14
707
0.14
ENDER
0.14
Activations Density 0.018%