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

    Neuronpedia

    INDEX
    Explanations

    function

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    inion
    -0.07
    ар
    -0.06
    zo
    -0.06
    ília
    -0.06
     Caesar
    -0.06
    apo
    -0.06
    (src
    -0.06
    argar
    -0.06
     breeding
    -0.06
    relationship
    -0.06
    POSITIVE LOGITS
    /react
    0.07
    _gp
    0.07
    PACK
    0.07
     owned
    0.07
     functions
    0.06
     Hardy
    0.06
    ookeeper
    0.06
     s
    0.06
     generosity
    0.06
     strconv
    0.06
    Act Density 0.004%

    No Known Activations