INDEX
Explanations
numeric values and age references in the text
New Auto-Interp
Negative Logits
ati
-0.16
ker
-0.16
ogan
-0.16
ab
-0.15
umi
-0.15
chez
-0.15
etc
-0.15
829
-0.15
ami
-0.14
ori
-0.14
POSITIVE LOGITS
edBy
0.17
agnar
0.17
éĹ
0.16
TextAlign
0.16
apart
0.15
alive
0.15
ìł¸
0.14
times
0.14
/close
0.14
å¯
0.14
Activations Density 0.188%