INDEX
Explanations
mentions of the Pew Research Center in the text
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
370
+0.20
0.9%
32
+0.18
0.9%
1363
+0.14
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
370
+0.20
0.04
981
+0.18
0.06
1363
+0.14
0.04
Negative Logits
Kuli
-0.59
tagHelperRunner
-0.54
granada
-0.53
Coc
-0.52
Cár
-0.52
principalTable
-0.51
Taro
-0.51
osse
-0.49
Coc
-0.49
testify
-0.49
POSITIVE LOGITS
Cav
0.66
Cav
0.61
Craig
0.61
Pew
0.61
Craig
0.57
dichi
0.54
utilizza
0.54
Campionato
0.53
Pew
0.53
svolge
0.53
Activations Density 0.331%