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

    Neuronpedia

    INDEX
    Explanations

    generating responses

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    notice
    0.62
     Notice
    0.58
    Notice
    0.57
     quits
    0.56
     виды
    0.55
     notice
    0.55
    Stewart
    0.55
     Sobolev
    0.55
     Sebasti
    0.54
     NOTICE
    0.54
    POSITIVE LOGITS
     paddock
    0.65
    腌
    0.63
     ಚೆ
    0.62
    glied
    0.61
    荒
    0.61
     chăm
    0.61
    ularis
    0.61
     fromage
    0.59
    鸠
    0.59
    囝
    0.58
    Act Density 0.000%

    No Known Activations