INDEX
Explanations
contextual markers indicating actions or events in a narrative
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
232
+0.13
0.7%
223
+0.12
0.7%
188
+0.12
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
223
+0.13
0.02
475
+0.12
0.02
420
+0.12
0.03
Negative Logits
OPINION
-1.61
ipl
-1.55
pic
-1.44
rapeut
-1.40
lli
-1.39
PH
-1.35
phon
-1.34
Kick
-1.31
blr
-1.30
/@
-1.30
POSITIVE LOGITS
ľ
3.23
ŀ
3.22
Ķ
3.19
¢
3.16
ļ
3.07
Ń
3.03
¨
3.01
ª
2.97
´
2.95
ĥ
2.95
Activations Density 0.331%