INDEX
Explanations
measurements in kilometers and miles
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1607
+0.07
0.2%
836
+0.07
0.2%
1486
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
181
+0.07
0.02
836
+0.07
0.01
1607
+0.07
0.02
Negative Logits
shenan
-1.38
unspeak
-1.27
apprehen
-1.18
pamph
-1.18
depic
-1.16
maneu
-1.13
indestru
-1.11
intersper
-1.11
resear
-1.10
philanth
-1.07
POSITIVE LOGITS
esigenze
0.61
utebol
0.59
Walkover
0.59
depoz
0.59
sandalias
0.58
beginnetje
0.58
órmula
0.58
relazioni
0.58
noten
0.58
conseguenze
0.57
Activations Density 0.013%