INDEX
Explanations
mentions of a specific name "Tak"
repeated mentions of the name "Tak"
New Auto-Interp
Negative Logits
DRAG
-0.73
anwhile
-0.72
ORPG
-0.71
pherd
-0.70
д
-0.66
livest
-0.66
reception
-0.66
theless
-0.64
guiActiveUnfocused
-0.64
circ
-0.61
POSITIVE LOGITS
Tak
1.17
umar
1.02
atsuki
1.01
amaru
1.00
una
0.98
istani
0.98
ashi
0.92
ota
0.92
ata
0.91
ita
0.91
Activations Density 0.005%