INDEX
Explanations
phrases with a positive connotation related to aesthetics
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
376
+0.12
0.3%
517
+0.10
0.3%
674
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
376
+0.12
0.07
554
+0.10
0.06
517
+0.10
0.03
Negative Logits
silikon
-1.06
kompres
-1.05
optik
-1.01
mikrofon
-0.99
akut
-0.93
karton
-0.92
keramik
-0.91
kafe
-0.91
alkoh
-0.86
katastro
-0.86
POSITIVE LOGITS
viewWillAppear
0.70
LOOKS
0.67
Genau
0.66
looked
0.65
getAge
0.65
viewWillAppear
0.64
sappi
0.62
Ottimo
0.60
looks
0.60
getAmount
0.59
Activations Density 0.155%