INDEX
Explanations
references to various countries, with a specific focus on Italy
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.12
0.7%
341
+0.11
0.6%
860
+0.11
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
860
+0.12
0.03
517
+0.11
0.03
131
+0.11
0.03
Negative Logits
<bos>
-2.57
Allora
-0.81
ⓧ
-0.72
Quindi
-0.63
Rispondi
-0.62
prevent
-0.61
since
-0.61
Tutto
-0.61
maximize
-0.60
Inoltre
-0.60
POSITIVE LOGITS
aen
1.72
ftu
1.60
lele
1.57
fta
1.55
thut
1.54
jaya
1.50
tew
1.48
myn
1.48
vns
1.48
mef
1.47
Activations Density 0.169%