INDEX
Explanations
occurrences of the word "onto."
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.23
1.3%
204
+0.19
1.1%
662
+0.14
0.8%
Correlated Neurons
Index
P. Corr.
Cos Sim.
204
+0.23
0.03
662
+0.19
0.02
1334
+0.14
0.02
Negative Logits
<bos>
-2.10
/**
-0.78
ⓧ
-0.77
<?
-0.75
deinit
-0.61
model
-0.59
qDebug
-0.59
Completo
-0.58
bar
-0.58
/***
-0.57
POSITIVE LOGITS
onto
1.14
Hæ
1.11
Czechos
1.10
Frö
1.07
maroc
1.07
saar
1.07
Græ
1.07
Juf
1.06
frankfurt
1.05
bayern
1.05
Activations Density 0.036%