INDEX
Explanations
descriptions of locations, attractions, and activities in a tourist destination
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.28
1.3%
1385
+0.14
0.7%
906
+0.12
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1120
+0.28
0.04
906
+0.14
0.01
1511
+0.12
0.03
Negative Logits
<bos>
-2.33
ⓧ
-0.90
<?
-0.86
-0.84
/**
-0.77
/*
-0.75
/***
-0.74
<?
-0.66
#
-0.63
fektions
-0.61
POSITIVE LOGITS
véhic
1.24
soulign
1.23
délib
1.19
ecru
1.15
prét
1.13
swarovski
1.12
épu
1.11
malheureux
1.10
écout
1.09
embodi
1.09
Activations Density 2.973%