INDEX
Explanations
phrases related to urban population growth and energy transition
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.29
1.1%
1253
+0.15
0.5%
764
+0.10
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1253
+0.29
0.04
609
+0.15
0.05
1484
+0.10
0.04
Negative Logits
<bos>
-2.38
ⓧ
-1.02
intersper
-0.91
-0.85
<?
-0.83
/***
-0.76
/**
-0.74
<?
-0.74
darted
-0.68
quitted
-0.66
POSITIVE LOGITS
bandung
1.00
maroc
0.90
jawa
0.90
kafe
0.87
nikah
0.86
kela
0.85
jaya
0.85
jati
0.84
surabaya
0.84
Informações
0.83
Activations Density 0.548%