INDEX
Explanations
possessive pronouns denoting ownership or connection to someone or something
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1385
+0.15
0.5%
1741
+0.10
0.3%
1978
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1317
+0.15
0.05
817
+0.10
0.03
1937
+0.10
0.04
Negative Logits
reluct
-1.90
snoopy
-1.88
milf
-1.88
impra
-1.87
affor
-1.86
scrat
-1.86
increa
-1.85
shenan
-1.82
disagre
-1.82
unspeak
-1.80
POSITIVE LOGITS
windowFixed
0.73
يميديا
0.72
insuffisamment
0.71
RTSC
0.66
<bos>
0.65
URLException
0.63
chest
0.63
Nº
0.61
HostException
0.60
ждународ
0.60
Activations Density 0.213%