INDEX
Explanations
repetitions of the digit '1' throughout the text
New Auto-Interp
Negative Logits
agra
-0.18
ůj
-0.17
ilen
-0.17
Tham
-0.17
entifier
-0.16
SCALL
-0.15
oa
-0.15
važ
-0.15
uji
-0.15
.Fields
-0.14
POSITIVE LOGITS
deb
0.17
ver
0.14
es
0.14
ewriter
0.14
christ
0.14
poly
0.14
losure
0.13
peter
0.13
persona
0.13
pe
0.13
Activations Density 0.003%