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

    Neuronpedia

    INDEX
    Explanations

    /b

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     compan
    -0.07
    .link
    -0.06
    =context
    -0.06
    であ
    -0.06
     şun
    -0.06
    _mail
    -0.06
     Geneva
    -0.06
     ngôn
    -0.06
    _booking
    -0.06
     antiqu
    -0.05
    POSITIVE LOGITS
     broadcaster
    0.07
    ジ
    0.07
    _fmt
    0.06
    jiště
    0.06
     nimi
    0.06
    music
    0.06
     inevitably
    0.06
    pr
    0.06
    립
    0.06
     살
    0.06
    Act Density 0.006%

    No Known Activations