INDEX
Explanations
information related to financial activities and economic outcomes
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
184
+0.19
0.6%
381
+0.15
0.5%
50
+0.12
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
184
+0.19
0.03
1965
+0.15
0.05
1577
+0.12
0.07
Negative Logits
morire
-0.69
berlina
-0.64
abbra
-0.59
mangiare
-0.58
venuto
-0.57
credere
-0.56
membrance
-0.56
adulti
-0.55
sposa
-0.55
churrasco
-0.54
POSITIVE LOGITS
depic
0.99
thereupon
0.96
intrigu
0.95
maneu
0.95
gaily
0.92
apprehen
0.91
unspeak
0.91
shenan
0.91
fortn
0.89
reconno
0.88
Activations Density 0.458%