INDEX
Explanations
phrases related to military or armed forces
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
812
+0.19
0.8%
1937
+0.17
0.7%
805
+0.14
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
812
+0.19
0.04
805
+0.17
0.03
1937
+0.14
0.03
Negative Logits
affez
-0.55
felice
-0.49
rejet
-0.49
kath
-0.48
trovo
-0.48
poff
-0.48
stihl
-0.48
fuf
-0.47
Persians
-0.47
Catal
-0.46
POSITIVE LOGITS
army
1.10
Army
1.08
Army
1.08
army
0.97
ARMY
0.84
armies
0.77
Armee
0.62
soldiers
0.62
Româ
0.59
ercito
0.58
Activations Density 0.052%