INDEX
Explanations
phrases related to ongoing challenges or movements
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
674
+0.10
0.3%
1986
+0.07
0.2%
1225
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
838
+0.10
0.03
97
+0.07
0.03
488
+0.07
0.03
Negative Logits
meis
-0.75
wald
-0.73
discogs
-0.72
aen
-0.71
wien
-0.69
inder
-0.67
keram
-0.65
geforce
-0.65
tille
-0.64
wein
-0.64
POSITIVE LOGITS
<bos>
0.72
EndProject
0.55
WebElementEntity
0.54
yet
0.53
Aún
0.52
Empieza
0.51
buquerque
0.50
DisplayMetrics
0.48
AndEndTag
0.48
endforeach
0.48
Activations Density 0.145%