INDEX
Explanations
words related to time and simultaneity
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.18
0.7%
814
+0.12
0.5%
1678
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
814
+0.18
0.04
333
+0.12
0.03
645
+0.11
0.03
Negative Logits
<bos>
-2.86
intersper
-0.91
endow
-0.75
ascribe
-0.74
darted
-0.74
enlist
-0.73
ratify
-0.72
mustered
-0.71
strove
-0.70
leapt
-0.70
POSITIVE LOGITS
Simult
1.06
cioc
1.06
sappi
1.01
maroc
1.01
meis
1.01
parati
0.99
ceramica
0.98
wien
0.98
ados
0.98
italia
0.97
Activations Density 0.166%