INDEX
Explanations
phrases related to natural elements or resources
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.17
1.0%
1671
+0.13
0.7%
1828
+0.08
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1671
+0.17
0.04
1370
+0.13
0.04
1575
+0.08
0.03
Negative Logits
<bos>
-3.35
ⓧ
-0.83
-0.77
add
-0.69
expand
-0.67
have
-0.66
invest
-0.66
-0.66
add
-0.65
spend
-0.65
POSITIVE LOGITS
Juf
1.76
wien
1.71
sappi
1.70
stockholm
1.70
napoli
1.69
maneu
1.68
aen
1.64
milano
1.60
thut
1.59
fua
1.58
Activations Density 0.077%