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

    Neuronpedia

    INDEX
    Explanations

    coverage

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     additionally
    -0.07
    /release
    -0.07
     Import
    -0.07
    .Of
    -0.06
    取消
    -0.06
    _origin
    -0.06
     parole
    -0.06
    convert
    -0.06
    _clause
    -0.06
    ящ
    -0.06
    POSITIVE LOGITS
    藍
    0.06
    가능
    0.06
     fakt
    0.06
    PLIC
    0.06
    labilir
    0.06
     gauche
    0.06
    kuk
    0.06
     Μέ
    0.06
    班
    0.06
    OOD
    0.06
    Act Density 0.005%

    No Known Activations