INDEX
Explanations
verbs related to removing or eliminating something, with a focus on erasing actions
the concept of erasure or removal
New Auto-Interp
Negative Logits
anooga
-0.97
annis
-0.84
uay
-0.83
razen
-0.75
liv
-0.71
oké
-0.71
regulated
-0.70
ombat
-0.68
enegger
-0.67
akov
-0.67
POSITIVE LOGITS
erase
1.34
erased
0.99
Doodle
0.83
overwrite
0.79
brush
0.79
Jagu
0.75
ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
0.74
ç¥ŀ
0.73
mson
0.71
marker
0.71
Activations Density 0.005%