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

    Neuronpedia

    INDEX
    Explanations

    'tis, til, cause, stats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    0.90
    া
    0.86
    ো
    0.85
    ।
    0.84
    ه
    0.82
    та
    0.81
    2
    0.81
    ט
    0.79
    "]
    0.77
    3
    0.75
    POSITIVE LOGITS
    يك
    0.65
    deki
    0.65
    ché
    0.61
     międzynarod
    0.60
    しくは
    0.60
    们的
    0.58
    giveness
    0.55
    Α
    0.55
    겅
    0.54
    Down
    0.54
    Act Density 0.019%

    No Known Activations