INDEX
Explanations
references to technology products or features, such as apps, versions, or devices
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1967
+0.14
0.4%
674
+0.14
0.4%
764
+0.13
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1995
+0.14
0.04
801
+0.14
0.04
1967
+0.13
0.04
Negative Logits
meis
-0.66
karton
-0.64
WebElementEntity
-0.64
printStats
-0.64
fometimes
-0.63
smtplib
-0.63
liev
-0.62
ftu
-0.60
silikon
-0.60
kado
-0.60
POSITIVE LOGITS
similarly
0.75
likewise
0.63
elsewhere
0.58
similar
0.57
nearby
0.57
comparable
0.55
ebenfalls
0.54
equally
0.54
conversely
0.52
theirs
0.51
Activations Density 0.748%