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

    Neuronpedia

    INDEX
    Explanations

    time horizons and durations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     week
    1.98
     month
    1.83
     weeks
    1.78
    week
    1.74
     semanas
    1.71
     semaines
    1.69
    Week
    1.68
     semana
    1.67
     WEEK
    1.67
     Month
    1.67
    POSITIVE LOGITS
     decade
    1.43
    decade
    1.18
    五年
    1.18
     Decade
    1.15
    十年
    1.12
     quinqu
    1.08
     década
    1.01
     пяти
    0.97
    five
    0.95
     dek
    0.95
    Act Density 0.176%

    No Known Activations