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

    Neuronpedia

    INDEX
    Explanations

    modelopen-source model characteristics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Clas
    0.48
    古
    0.46
    如今
    0.45
    Black
    0.44
    首先
    0.44
     alternativa
    0.43
     aufge
    0.43
    стьян
    0.43
    琰
    0.43
    Ol
    0.43
    POSITIVE LOGITS
    od
    0.54
    ot
    0.53
    ate
    0.53
    ud
    0.52
     تیز
    0.50
    oną
    0.50
    ite
    0.49
    indice
    0.49
    ia
    0.48
    update
    0.48
    Act Density 0.001%

    No Known Activations