INDEX
Explanations
mentions of a specific country, Brazil
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1507
+0.14
0.5%
2004
+0.12
0.4%
1331
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1507
+0.14
0.03
227
+0.12
0.04
395
+0.11
0.03
Negative Logits
Áng
-1.36
Valentín
-1.33
Darío
-1.32
Água
-1.31
Compañ
-1.27
Lázaro
-1.24
Vitória
-1.22
Mulher
-1.21
Belén
-1.19
Cár
-1.17
POSITIVE LOGITS
Braz
1.22
Brazilian
1.18
Brazil
1.14
brazil
1.13
Brazilian
1.12
Brazil
1.11
Brésil
1.08
brazilian
1.08
Braz
0.99
brazilian
0.96
Activations Density 0.189%