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

    Neuronpedia

    INDEX
    Explanations

    regularly do actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     erneut
    0.59
    再一次
    0.54
     opnieuw
    0.51
    再び
    0.50
    这将
    0.50
    再度
    0.48
     ponownie
    0.47
     மீண்டும்
    0.47
     প্রথমবারের
    0.46
    again
    0.46
    POSITIVE LOGITS
     regularly
    1.01
     regelmäßig
    0.95
     regularmente
    0.94
     sometimes
    0.92
     habituellement
    0.92
     регулярно
    0.92
     occasionally
    0.89
     иногда
    0.89
     주로
    0.89
     regelmatig
    0.88
    Act Density 0.025%

    No Known Activations