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

    Neuronpedia

    INDEX
    Explanations

    code and technical discussions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     purposes
    -0.07
     registrations
    -0.06
    .click
    -0.06
     Mehr
    -0.06
     Silence
    -0.06
    导致
    -0.06
     Москва
    -0.06
    ocols
    -0.06
     βρί
    -0.06
    Summary
    -0.06
    POSITIVE LOGITS
    .robot
    0.07
    wp
    0.07
     lParam
    0.07
    icularly
    0.06
    essenger
    0.06
    /views
    0.06
    (e
    0.06
     Comcast
    0.06
    547
    0.06
     Formatting
    0.06
    Act Density 0.000%

    No Known Activations