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

    Neuronpedia

    INDEX
    Explanations

    open weights

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    మ్
    0.44
    ίων
    0.43
    น
    0.41
    ือ
    0.41
    про
    0.40
    Solve
    0.40
    עת
    0.40
    ำ
    0.40
    "
    0.39
    ดู
    0.39
    POSITIVE LOGITS
     editar
    0.53
    ۵
    0.52
    <unused1016>
    0.50
    选取
    0.48
     escolar
    0.48
     confocal
    0.48
     පේශ
    0.48
    cycled
    0.47
     ڕ
    0.47
     delanter
    0.47
    Act Density 0.011%

    No Known Activations