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

    Neuronpedia

    INDEX
    Explanations

    JavaScript, angiotensin, Korean

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     كل
    0.48
    weighted
    0.46
     ور
    0.45
    غي
    0.44
     سه
    0.44
     жал
    0.43
    parameter
    0.43
     einfacher
    0.43
    ر
    0.42
    ل
    0.42
    POSITIVE LOGITS
    larini
    0.53
     doesnt
    0.49
     prone
    0.47
     pupp
    0.47
     susceptible
    0.46
    ள்ளதாக
    0.46
     relay
    0.45
    !,
    0.45
     obi
    0.45
     purporting
    0.45
    Act Density 0.001%

    No Known Activations