INDEX
Explanations
occurrences of the character "í"
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
493
+0.13
0.7%
9
+0.13
0.7%
308
+0.12
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
111
+0.13
0.02
7
+0.13
0.01
214
+0.12
0.01
Negative Logits
uties
-1.76
ce
-1.56
ças
-1.55
pat
-1.42
ception
-1.41
cep
-1.39
amus
-1.38
thereon
-1.37
ucle
-1.36
ceptor
-1.36
POSITIVE LOGITS
Ĵ
3.95
ĺ
3.70
Į
3.59
ĥ
3.59
İ
3.53
Ŀ
3.50
ŀ
3.49
Ľ
3.48
Ļ
3.44
ı
3.41
Activations Density 0.030%