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

    Neuronpedia

    INDEX
    Explanations

    Further explanation/details

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (no
    -0.07
     Desert
    -0.06
    Titan
    -0.06
    Sea
    -0.06
     Cool
    -0.06
     dislikes
    -0.06
    ,class
    -0.06
     Breath
    -0.06
    .Perform
    -0.06
     Sea
    -0.06
    POSITIVE LOGITS
     tyto
    0.07
     wnd
    0.06
    olicy
    0.06
    subst
    0.06
    entral
    0.06
    -th
    0.06
    ested
    0.06
     retrospective
    0.06
     pend
    0.06
    .setVisibility
    0.06
    Act Density 0.048%

    No Known Activations