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

    Neuronpedia

    INDEX
    Explanations

    occasional events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    poi
    0.50
     aussi
    0.45
    champ
    0.45
    och
    0.43
    paras
    0.42
    prettier
    0.42
    treasure
    0.42
    transform
    0.42
    ta
    0.41
    pera
    0.41
    POSITIVE LOGITS
     kadang
    0.55
     timedelta
    0.45
     soms
    0.45
    بعض
    0.44
     بعض
    0.44
    一些
    0.43
     නම්
    0.43
     sometimes
    0.43
     কিছু
    0.40
    ंसू
    0.40
    Act Density 0.053%

    No Known Activations