INDEX
Explanations
helper verbs signaling analysis or action, as well as technological terms related to software and systems operation
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1385
+0.09
0.3%
478
+0.09
0.3%
876
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1543
+0.09
0.04
369
+0.09
0.04
1398
+0.08
0.03
Negative Logits
unspeak
-1.12
gaily
-1.12
McLaugh
-1.06
reluct
-0.99
ineffec
-0.99
indescri
-0.98
encomp
-0.97
apprehen
-0.96
nobly
-0.94
shenan
-0.94
POSITIVE LOGITS
distanciation
0.79
vola
0.71
<bos>
0.71
gamba
0.70
pompa
0.70
siasme
0.69
solidar
0.69
reputa
0.69
ideolog
0.69
ształ
0.69
Activations Density 0.343%