INDEX
Explanations
information related to event durations stated in days or years
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
453
+0.17
0.6%
1535
+0.17
0.5%
1967
+0.13
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
297
+0.17
0.04
1884
+0.17
0.03
446
+0.13
0.03
Negative Logits
indescri
-0.96
unspeak
-0.95
perchance
-0.84
intersper
-0.80
horrend
-0.80
plenti
-0.77
unavoid
-0.77
miroir
-0.77
exult
-0.76
vainly
-0.76
POSITIVE LOGITS
silikon
1.32
solidar
1.30
alkoh
1.23
kosme
1.22
utop
1.20
ideolog
1.18
keramik
1.18
praktik
1.18
kafe
1.18
demokra
1.16
Activations Density 0.110%