INDEX
Explanations
words that convey an essential or fundamental quality or action
New Auto-Interp
Negative Logits
æķ·
-0.20
ycz
-0.16
unsch
-0.15
jdk
-0.14
gens
-0.14
orent
-0.14
rous
-0.14
edin
-0.14
oret
-0.14
تاب
-0.14
POSITIVE LOGITS
daylight
0.14
forb
0.14
amus
0.14
779
0.14
spoilers
0.14
κι
0.14
Durant
0.13
essentially
0.13
ionate
0.13
darwin
0.13
Activations Density 0.043%