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

    Neuronpedia

    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     tokenize
    -0.07
     grilled
    -0.07
     believe
    -0.07
     hailed
    -0.07
    [W
    -0.07
     bao
    -0.06
    descending
    -0.06
    qli
    -0.06
     believes
    -0.06
     furnish
    -0.06
    POSITIVE LOGITS
     dignity
    0.06
    =======
    0.06
    出版社
    0.06
    olk
    0.06
     Dillon
    0.06
    _UNITS
    0.06
    ,sizeof
    0.05
     Packs
    0.05
    achat
    0.05
     Denver
    0.05
    Act Density 0.055%

    No Known Activations