INDEX
Explanations
"I can", "I have" constructs
New Auto-Interp
Negative Logits
politicians
0.49
studying
0.46
cout
0.43
America
0.43
drawbacks
0.42
ixed
0.42
home
0.41
advantage
0.41
Africa
0.41
Politicians
0.41
POSITIVE LOGITS
c
0.55
屠
0.54
quando
0.53
}+{\0.53
ifadə
0.52
koliko
0.50
ﺩ
0.49
mantenere
0.49
formul
0.48
adamu
0.48
Activations Density 0.001%