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

    Neuronpedia

    INDEX
    Explanations

    tricky to

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     фактически
    0.97
    おそらく
    0.84
    Mostly
    0.81
     Mostly
    0.77
     mainly
    0.76
     hauptsächlich
    0.75
     principalement
    0.75
     currently
    0.74
     primarily
    0.74
     principally
    0.73
    POSITIVE LOGITS
     бывает
    1.33
     sometimes
    1.11
    Sometimes
    1.05
    こういう
    1.04
     always
    1.03
    sometimes
    1.02
     таке
    1.00
     sowas
    1.00
     altijd
    0.99
     rarely
    0.98
    Act Density 0.562%

    No Known Activations