INDEX
Explanations
objects that are spherical or circular in shape
New Auto-Interp
Negative Logits
nem
-0.15
772
-0.14
ÑĥÑĩаÑģÑĤи
-0.14
folded
-0.14
lines
-0.14
ouble
-0.13
亿
-0.13
á
-0.13
à¤ļल
-0.13
Peg
-0.13
POSITIVE LOGITS
-shaped
0.19
ặn
0.18
atform
0.16
warts
0.16
sono
0.15
stance
0.14
uales
0.14
Äįast
0.14
ois
0.14
Heard
0.14
Activations Density 0.213%