INDEX
Explanations
words related to emotional expressions or feelings
New Auto-Interp
Negative Logits
addtogroup
-0.17
ноз
-0.16
qui
-0.14
cred
-0.14
raid
-0.14
ORIGINAL
-0.14
ÑĢÑĥж
-0.13
аков
-0.13
FullScreen
-0.13
ÅĪ
-0.13
POSITIVE LOGITS
note
0.21
note
0.20
Note
0.17
">//
0.17
Note
0.17
Pig
0.17
-note
0.16
ìĤ¬íķŃ
0.16
notes
0.15
å¤ĩ注
0.15
Activations Density 0.010%