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

    Neuronpedia

    INDEX
    Explanations

    code pivot assignment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    क्सी
    0.48
    傈
    0.41
    лли
    0.40
    સુ
    0.40
    鬏
    0.40
    ಧಾರವಾಡ
    0.39
     chicas
    0.38
    фи
    0.37
    сси
    0.37
     riguarda
    0.37
    POSITIVE LOGITS
     sẽ
    0.47
    จะเป็น
    0.46
    communicate
    0.42
     sending
    0.41
     will
    0.41
     θα
    0.39
     envoyer
    0.39
     रहेगी
    0.39
    asku
    0.38
     communicate
    0.38
    Act Density 0.001%

    No Known Activations