INDEX
Explanations
special characters or a specific character pattern "ķ"
specific characters or symbols in text
New Auto-Interp
Negative Logits
arios
-0.77
iflower
-0.68
iasm
-0.68
gestation
-0.66
iewicz
-0.65
narrowly
-0.64
Nadu
-0.64
apprentice
-0.64
Aber
-0.64
aic
-0.63
POSITIVE LOGITS
¾
0.89
Ķ
0.88
vernment
0.87
reg
0.83
press
0.82
Column
0.80
flush
0.80
raise
0.80
ij
0.79
λ
0.79
Activations Density 0.009%