INDEX
Explanations
preposition phrases starting with 'in' and 'a'
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.14
0.6%
390
+0.11
0.5%
1506
+0.10
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1539
+0.14
0.04
507
+0.11
0.03
1577
+0.10
0.07
Negative Logits
<bos>
-2.74
/***
-1.04
ⓧ
-1.00
///**
-0.91
//*/
-0.81
-0.81
<?
-0.79
<?
-0.78
تفصیلات
-0.76
/**
-0.74
POSITIVE LOGITS
lccccc
0.67
teflon
0.66
:"-
0.63
tupperware
0.63
softshell
0.60
PTFE
0.59
-
0.54
lcccccc
0.54
hilux
0.53
lcccc
0.52
Activations Density 0.697%