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

    Neuronpedia

    INDEX
    Explanations

    hit

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     catastrophic
    -0.07
    Fox
    -0.07
    Escape
    -0.06
    三年
    -0.06
     больш
    -0.06
    orris
    -0.06
     wasn
    -0.06
     %"
    -0.06
    forth
    -0.06
    aac
    -0.06
    POSITIVE LOGITS
     Or
    0.07
     operative
    0.06
    _DS
    0.06
    地區
    0.06
     giy
    0.06
    /testing
    0.06
    _dense
    0.06
    aintenance
    0.06
     Mandarin
    0.06
     TASK
    0.06
    Act Density 0.015%

    No Known Activations