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

    Neuronpedia

    INDEX
    Explanations

    legal code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    endant
    -0.08
    .xr
    -0.07
    _are
    -0.07
     shades
    -0.07
     famille
    -0.07
     σ
    -0.06
    dG
    -0.06
    theta
    -0.06
    swana
    -0.06
    _l
    -0.06
    POSITIVE LOGITS
    .previous
    0.07
    (Command
    0.07
     CARE
    0.06
    プリ
    0.06
     перший
    0.06
    .groupControl
    0.06
     pravděpodob
    0.06
    タイ
    0.06
    叔
    0.06
     unavailable
    0.06
    Act Density 0.008%

    No Known Activations