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

    Neuronpedia

    INDEX
    Explanations

    that can

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     the
    0.53
    ating
    0.50
     this
    0.46
     unhealthy
    0.45
    р
    0.43
    ва
    0.43
     quer
    0.42
     their
    0.41
     consegu
    0.41
     nhật
    0.41
    POSITIVE LOGITS
     Außerdem
    0.51
    can
    0.48
    não
    0.48
    mim
    0.47
     iż
    0.46
     می‌تواند
    0.44
     يساعد
    0.44
     sustains
    0.44
     specializes
    0.43
     može
    0.43
    Act Density 0.158%

    No Known Activations