INDEX
Explanations
key features and selling points of a product or service, possibly related to pets or home decor
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.25
1.7%
478
+0.12
0.8%
1937
+0.11
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1937
+0.25
0.08
131
+0.12
0.07
144
+0.11
0.07
Negative Logits
<bos>
-2.93
/**
-1.18
ⓧ
-0.92
/*
-0.86
-0.82
<?
-0.79
<?
-0.77
/***
-0.73
///**
-0.69
<!--
-0.67
POSITIVE LOGITS
véhic
0.92
Our
0.89
Our
0.88
Juf
0.87
Khart
0.86
Pekan
0.86
Punj
0.86
Keny
0.84
réput
0.84
our
0.82
Activations Density 0.205%