Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    Finn and character actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kullan
    0.86
     사용하는
    0.68
     verwenden
    0.67
     kullanımı
    0.66
     kullanım
    0.66
     doit
    0.66
     kullanıl
    0.66
     അല്ലെങ്കിൽ
    0.65
    を使用
    0.65
    사용
    0.63
    POSITIVE LOGITS
    didn
    0.75
     responded
    0.73
    последствии
    0.71
     warned
    0.70
     knew
    0.70
     refused
    0.69
     sighed
    0.68
    had
    0.68
     insisted
    0.68
     dismay
    0.67
    Act Density 0.105%

    No Known Activations