INDEX
Explanations
No Explanations Found
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
674
+0.38
2.3%
1741
+0.07
0.5%
1403
+0.04
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
2
-0.38
0.00
0
-0.07
0.00
1
-0.04
0.00
Negative Logits
belliger
-1.62
ruinous
-1.50
unlaw
-1.46
unspeak
-1.45
ineffec
-1.43
despotism
-1.38
nukes
-1.38
impractica
-1.37
horrend
-1.36
miscon
-1.35
POSITIVE LOGITS
<bos>
12.89
GEBURTSDATUM
2.56
expandindo
2.53
betweenstory
2.41
Autoritní
2.35
kasarigan
2.23
تقاوى
2.10
Administrativna
2.08
LookAnd
2.05
Italijani
2.04
Activations Density 0.000%
No Known Activations
This feature has no known activations.