INDEX
Explanations
sentences discussing financial and business matters
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
604
+0.09
0.3%
1531
+0.08
0.2%
651
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1232
+0.09
0.04
1531
+0.08
0.04
1791
+0.08
0.04
Negative Logits
<bos>
-0.95
fta
-0.77
mme
-0.76
wien
-0.73
mef
-0.72
emphat
-0.72
afo
-0.71
idem
-0.70
nikah
-0.70
vnd
-0.69
POSITIVE LOGITS
ercice
0.68
only
0.64
Біографія
0.62
Lmfao
0.58
Бележки
0.58
Життєпис
0.58
XmlEnum
0.58
Áng
0.57
">/
0.56
Ehh
0.55
Activations Density 0.402%