INDEX
Explanations
events or accidents involving physical impact or danger
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
906
+0.15
0.4%
509
+0.11
0.3%
1385
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
736
+0.15
0.06
509
+0.11
0.05
1644
+0.10
0.02
Negative Logits
minimalis
-1.25
alkoh
-1.18
praktik
-1.17
kosme
-1.16
silikon
-1.14
stoff
-1.13
pól
-1.13
biograf
-1.11
krab
-1.07
kram
-1.06
POSITIVE LOGITS
thereupon
0.76
disreg
0.75
impelled
0.71
philosophic
0.68
felicity
0.67
quitted
0.66
vicissitudes
0.66
adjour
0.66
subgoals
0.66
impractica
0.65
Activations Density 0.229%