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

    Neuronpedia

    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     sommes
    -0.07
     histo
    -0.07
    魂
    -0.07
     tobacco
    -0.07
    habi
    -0.07
     decimal
    -0.07
    .nil
    -0.07
    argins
    -0.07
    -stop
    -0.06
    üp
    -0.06
    POSITIVE LOGITS
    []={
    0.06
    Immediately
    0.06
    _FRAGMENT
    0.06
    TITLE
    0.06
     タ
    0.06
     iii
    0.05
     acess
    0.05
    (ws
    0.05
    legation
    0.05
     частина
    0.05
    Act Density 0.034%

    No Known Activations