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

    Neuronpedia

    INDEX
    Explanations

    Code/API

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     attended
    -0.08
     Crime
    -0.07
    .Values
    -0.07
    DEC
    -0.06
     Feature
    -0.06
     Reed
    -0.06
     Vacuum
    -0.06
    telegram
    -0.06
     Thing
    -0.06
     aque
    -0.06
    POSITIVE LOGITS
     milf
    0.07
    (conf
    0.07
    (expr
    0.06
    祝
    0.06
    ythe
    0.06
     гот
    0.06
    eniable
    0.06
    [${
    0.06
     jamais
    0.06
    [..
    0.06
    Act Density 0.003%

    No Known Activations