INDEX
Explanations
phrases related to urgency and timeliness
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
2034
+0.13
0.4%
1013
+0.11
0.3%
1110
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
235
+0.13
0.05
284
+0.11
0.05
1013
+0.08
0.05
Negative Logits
Cześć
-0.74
Sklici
-0.71
Glej
-0.70
você
-0.70
Voyez
-0.69
honnête
-0.67
Bardzo
-0.67
Zunanje
-0.66
Dziękuję
-0.65
glLoad
-0.65
POSITIVE LOGITS
aneity
0.57
(<
0.55
kaos
0.55
optik
0.54
teater
0.53
akut
0.50
chaises
0.50
ïe
0.49
without
0.48
herbes
0.48
Activations Density 0.292%