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

    Neuronpedia

    INDEX
    Explanations

    etc

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ēi
    0.55
     ʻ
    0.54
     succinct
    0.53
    احث
    0.53
     ⚫
    0.52
     smoothies
    0.51
     plaus
    0.50
     smoothie
    0.50
     کارگرد
    0.50
     miscarriage
    0.50
    POSITIVE LOGITS
    etc
    0.64
     etc
    0.53
    ----------------
    0.48
    など
    0.47
    конец
    0.46
    var
    0.45
    ------------
    0.44
    EOF
    0.44
    --------------
    0.43
     തുടങ്ങി
    0.43
    Act Density 0.874%

    No Known Activations