INDEX
Explanations
words related to technology and updates in a system or software
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
752
+0.12
0.3%
678
+0.11
0.3%
690
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
752
+0.12
0.05
1322
+0.11
0.04
678
+0.10
0.06
Negative Logits
pecuni
-0.71
ביוגרפיה
-0.66
تانيه
-0.65
LookAnd
-0.65
girasol
-0.64
שוליים
-0.63
ohist
-0.62
ihnachten
-0.62
ercice
-0.62
quidem
-0.62
POSITIVE LOGITS
intersper
0.83
unavoid
0.77
Thier
0.73
maneu
0.72
sophistic
0.70
indescri
0.69
Opportun
0.69
every
0.68
confé
0.68
tolerably
0.68
Activations Density 0.468%