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

    Neuronpedia

    INDEX
    Explanations

    positive attributes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    authenticate
    -0.07
    Science
    -0.07
    𝒌
    -0.07
    医
    -0.07
    cmd
    -0.06
    łatw
    -0.06
    🌕
    -0.06
    𬍛
    -0.06
    נקוד
    -0.06
    .datasets
    -0.06
    POSITIVE LOGITS
    녔
    0.07
    を集
    0.07
    iros
    0.07
     giọng
    0.07
    .geom
    0.07
     Banner
    0.07
    atos
    0.07
    ario
    0.07
     Gron
    0.07
    ของเขา
    0.07
    Act Density 0.143%

    No Known Activations