INDEX
Explanations
themes of love and personal relationships
New Auto-Interp
Negative Logits
unes
-0.16
lobs
-0.15
.metro
-0.15
Ends
-0.14
ueva
-0.14
issy
-0.14
ogle
-0.14
_raises
-0.14
endar
-0.14
ender
-0.13
POSITIVE LOGITS
shines
0.27
shine
0.25
emerges
0.23
apparent
0.23
evident
0.21
rub
0.20
emerge
0.20
rubbing
0.20
rubbed
0.20
shining
0.20
Activations Density 0.139%