INDEX
Explanations
phrases related to financial investments and company development
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.12
0.6%
783
+0.12
0.6%
31
+0.11
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
783
+0.12
0.02
101
+0.12
0.02
690
+0.11
0.02
Negative Logits
<bos>
-2.45
ⓧ
-0.86
<?
-0.67
<?
-0.65
/*
-0.61
relieve
-0.60
Programa
-0.59
app
-0.59
Literatura
-0.58
↗
-0.57
POSITIVE LOGITS
increa
1.61
Confu
1.50
inev
1.47
unspeak
1.45
ecru
1.44
ftu
1.43
jorge
1.43
swarovski
1.40
Minang
1.40
wherea
1.39
Activations Density 0.190%