INDEX
Explanations
mentions of fatal incidents or tragedies
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1096
+0.13
0.4%
1618
+0.12
0.4%
1085
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
468
+0.13
0.03
765
+0.12
0.03
1618
+0.11
0.02
Negative Logits
Enllaces
-0.53
Brainz
-0.53
erk
-0.53
ManagedBean
-0.53
behar
-0.49
oneofs
-0.48
tami
-0.47
koc
-0.47
Gere
-0.47
kosme
-0.47
POSITIVE LOGITS
fatal
0.97
deadly
0.80
Fatal
0.79
lethal
0.78
fatally
0.74
Deadly
0.73
fatal
0.73
Deadly
0.71
dichiarato
0.64
lethal
0.64
Activations Density 0.091%