INDEX
Explanations
words related to the banking industry
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1870
+0.14
0.5%
1691
+0.11
0.4%
1092
+0.10
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1870
+0.14
0.04
1865
+0.11
0.05
196
+0.10
0.05
Negative Logits
reluct
-0.73
Sklici
-0.70
Glej
-0.63
bénéfices
-0.62
inappro
-0.61
dilap
-0.61
Barriers
-0.61
DropColumn
-0.60
unequiv
-0.60
disambigu
-0.60
POSITIVE LOGITS
ouvre
0.55
ouching
0.54
licensing
0.54
trist
0.53
strade
0.53
envoie
0.52
brille
0.52
fishing
0.52
unteering
0.51
gagne
0.51
Activations Density 0.364%