INDEX
Explanations
information related to landmarks and infrastructure
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
2034
+0.14
0.4%
1042
+0.13
0.4%
382
+0.11
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
382
+0.14
0.08
1445
+0.13
0.07
1042
+0.11
0.06
Negative Logits
franz
-0.97
pymysql
-0.95
ohr
-0.95
fup
-0.93
meis
-0.91
robus
-0.90
„,
-0.90
ftu
-0.89
paff
-0.87
kug
-0.86
POSITIVE LOGITS
surpassing
0.54
之一
0.50
Its
0.50
according
0.48
Highest
0.48
Yet
0.47
boasting
0.46
surpassed
0.46
Highest
0.45
undisputed
0.45
Activations Density 0.613%