INDEX
Explanations
names starting with "Eric" and potentially related information about those individuals
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
313
+0.14
0.5%
168
+0.11
0.4%
390
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
313
+0.14
0.02
168
+0.11
0.02
390
+0.11
0.02
Negative Logits
フォー
-0.47
shape
-0.47
表单
-0.46
државе
-0.45
الدع
-0.45
feature
-0.44
пусти
-0.44
فور
-0.44
organizacji
-0.43
depleted
-0.43
POSITIVE LOGITS
Eric
1.30
ERIC
1.28
Eric
1.26
alkoh
1.17
silikon
1.14
kask
1.14
praktik
1.11
eric
1.11
stoff
1.10
sappi
1.09
Activations Density 0.074%