INDEX
Explanations
references to political actions related to science and legal processes
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.34
1.4%
1577
+0.19
0.8%
1842
+0.15
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1577
+0.34
0.14
1385
+0.19
0.11
998
+0.15
0.07
Negative Logits
<bos>
-2.14
intersper
-1.13
encomp
-0.81
shenan
-0.80
chanced
-0.77
mustered
-0.77
/***
-0.76
endow
-0.74
enshr
-0.73
disambigu
-0.73
POSITIVE LOGITS
soggior
0.76
cioc
0.76
croce
0.74
/>";
0.73
ceramica
0.72
adatta
0.71
particolar
0.69
indiri
0.67
€/
0.67
mangiare
0.67
Activations Density 2.838%