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

    Neuronpedia

    INDEX
    Explanations

    page content classification

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ح
    0.59
    ли
    0.55
    ח
    0.55
    हा
    0.53
    hade
    0.46
     preval
    0.46
    ロ
    0.45
    死去
    0.45
    h
    0.45
     savory
    0.45
    POSITIVE LOGITS
    щение
    0.44
    defect
    0.41
    ransform
    0.41
     defect
    0.40
    asyon
    0.40
     consort
    0.40
     Fairness
    0.40
    SECOND
    0.40
    ρίου
    0.40
    ﾞ
    0.40
    Act Density 0.001%

    No Known Activations