INDEX
Explanations
the name "Takashi"
repeated mentions of a specific individual, particularly focusing on the name "Takashi."
New Auto-Interp
Negative Logits
count
-0.85
Count
-0.76
counts
-0.71
citizenship
-0.67
vector
-0.66
rect
-0.66
extrad
-0.64
PO
-0.63
Mathemat
-0.63
Count
-0.63
POSITIVE LOGITS
ashi
4.69
aru
1.41
uchi
1.22
ash
1.21
ushi
1.21
ataka
1.20
oshi
1.19
ishi
1.18
asu
1.18
ami
1.14
Activations Density 0.007%