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

    Neuronpedia

    INDEX
    Explanations

    question types and descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     answers
    0.40
     بدون
    0.39
     دون
    0.39
     المطل
    0.39
     bith
    0.39
     Diva
    0.37
     без
    0.37
    託
    0.36
    swers
    0.36
     cosm
    0.36
    POSITIVE LOGITS
    RQ
    0.57
    dq
    0.48
    cq
    0.48
     CQ
    0.46
    rq
    0.45
    qt
    0.45
    qr
    0.44
    TQ
    0.44
    CQ
    0.42
    QC
    0.42
    Act Density 0.001%

    No Known Activations