INDEX
Explanations
words conveying actions and states of being
New Auto-Interp
Negative Logits
borg
-0.15
MouseButton
-0.14
ije
-0.14
lip
-0.14
ica
-0.13
_DICT
-0.13
male
-0.13
896
-0.13
aurus
-0.13
xic
-0.13
POSITIVE LOGITS
anj
0.19
anje
0.15
aan
0.14
pinch
0.14
unan
0.14
wire
0.14
minus
0.14
aland
0.13
apr
0.13
CCR
0.13
Activations Density 0.009%