INDEX
Explanations
instances of the word "compare" and related discussions of comparison
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.33
1.5%
1387
+0.11
0.5%
1325
+0.11
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1981
+0.33
0.03
1387
+0.11
0.03
1325
+0.11
0.02
Negative Logits
<bos>
-2.19
/***
-0.72
ⓧ
-0.67
FlatAppearance
-0.64
fund
-0.63
enter
-0.57
enter
-0.57
/**
-0.56
Fund
-0.55
em
-0.55
POSITIVE LOGITS
aen
1.50
ftu
1.47
maneu
1.43
madonna
1.37
fta
1.36
ftre
1.35
affor
1.34
fatis
1.33
verona
1.33
thut
1.33
Activations Density 0.061%