INDEX
Explanations
No Explanations Found
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
674
+0.23
0.7%
1224
+0.05
0.2%
1870
+0.05
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
2
-0.23
0.00
0
-0.05
0.00
1
-0.05
0.00
Negative Logits
unspeak
-1.92
ruinous
-1.71
ineffec
-1.68
belliger
-1.67
miscon
-1.67
horrend
-1.63
vainly
-1.63
odious
-1.60
laug
-1.59
impelled
-1.59
POSITIVE LOGITS
<bos>
8.47
GEBURTSDATUM
2.13
expandindo
2.08
newswire
1.89
kasarigan
1.78
Autoritní
1.76
Personendaten
1.69
Paglinawan
1.64
Italijani
1.60
дописавши
1.60
Activations Density 0.000%
No Known Activations
This feature has no known activations.