INDEX
Explanations
words related to alarms and alarm systems
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.12
1.0%
1870
+0.05
0.4%
1385
+0.03
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1385
+0.12
0.19
1870
+0.05
0.03
1896
+0.03
0.05
Negative Logits
<bos>
-2.26
ⓧ
-0.89
-0.87
/**
-0.85
public
-0.83
/*
-0.82
<?
-0.80
,
-0.77
/**
-0.76
//
-0.76
POSITIVE LOGITS
affor
1.93
stockholm
1.90
Juf
1.89
aen
1.87
increa
1.84
maneu
1.82
fta
1.81
hcm
1.79
lidl
1.77
ftu
1.77
Activations Density 2.788%