INDEX
Explanations
instances of the emoji character
New Auto-Interp
Negative Logits
ÙĦØ·
-0.17
corner
-0.16
enna
-0.15
otte
-0.15
wan
-0.15
ä¿
-0.15
infra
-0.15
yg
-0.14
itsu
-0.14
corner
-0.14
POSITIVE LOGITS
Ń
0.22
¨
0.22
ĸ
0.22
Ī
0.19
©
0.18
®
0.17
Ĩ
0.17
§
0.16
¯u
0.16
ĩ
0.15
Activations Density 0.003%