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

    Neuronpedia

    INDEX
    Explanations

    math problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     hi
    -0.08
     sk
    -0.08
    434
    -0.08
    _SK
    -0.07
     cho
    -0.07
    DK
    -0.07
    usk
    -0.07
     Pentagon
    -0.07
    .demo
    -0.07
    .listeners
    -0.07
    POSITIVE LOGITS
    éin
    0.08
    оскольку
    0.07
     gab
    0.07
     candidates
    0.07
     geopolitical
    0.07
     Candidates
    0.07
    0.07
    елик
    0.07
    <<<<<<<
    0.07
    টু
    0.07
    Act Density 0.162%
    No Known Activations