INDEX
Explanations
No Explanations Found
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
674
+0.31
1.2%
1225
+0.05
0.2%
899
+0.05
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
2
-0.31
0.00
0
-0.05
0.00
1
-0.05
0.00
Negative Logits
unspeak
-2.64
reluct
-2.44
disagre
-2.28
shenan
-2.27
unlaw
-2.27
impra
-2.27
impractica
-2.26
unwarran
-2.22
ineffec
-2.17
indestru
-2.15
POSITIVE LOGITS
<bos>
10.07
Paglinawan
1.95
GEBURTSDATUM
1.91
expandindo
1.89
Autoritní
1.82
betweenstory
1.81
'\\;'
1.73
Walkover
1.71
Administrativna
1.68
تقاوى
1.65
Activations Density 0.000%
No Known Activations
This feature has no known activations.