INDEX
Explanations
mentions of the country of Turkey
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
544
+0.12
0.4%
1351
+0.11
0.4%
316
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
544
+0.12
0.03
1565
+0.11
0.03
316
+0.11
0.03
Negative Logits
tubercle
-0.80
ınır
-0.79
üyü
-0.69
papilla
-0.69
calyx
-0.68
carina
-0.67
abnorm
-0.67
Tja
-0.67
mondeo
-0.66
Samstag
-0.66
POSITIVE LOGITS
Turkey
1.29
Turkey
1.25
Turkish
1.20
Turks
1.08
Turkish
1.06
turkey
0.99
turco
0.97
Istanbul
0.90
Turk
0.86
Tur
0.85
Activations Density 0.128%