INDEX
Explanations
references to possession or ownership
New Auto-Interp
Negative Logits
aucoup
-0.16
ihn
-0.16
emb
-0.15
üs
-0.15
iotics
-0.14
ervlet
-0.14
zug
-0.14
ocus
-0.13
Added
-0.13
uze
-0.13
POSITIVE LOGITS
never
0.23
recently
0.19
never
0.18
been
0.18
just
0.16
nowhere
0.16
åĪļ
0.15
experience
0.15
lived
0.15
rippling
0.15
Activations Density 0.200%