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

    Neuronpedia

    INDEX
    Explanations

    code and results

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     loi
    0.55
     A
    0.49
     constant
    0.49
     sabbatical
    0.48
    甥
    0.48
     Weaver
    0.47
    arbij
    0.47
     violin
    0.47
     stitch
    0.47
     The
    0.46
    POSITIVE LOGITS
     результатов
    0.55
    notification
    0.50
    ata
    0.49
    that
    0.48
    money
    0.47
    im
    0.47
    Notification
    0.47
    Results
    0.46
    lement
    0.46
    covering
    0.46
    Act Density 0.000%

    No Known Activations