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

    Neuronpedia

    INDEX
    Explanations

    ast

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Based
    -0.09
    还算
    -0.09
    反映出
    -0.07
    inidad
    -0.07
    มา
    -0.07
     bảo
    -0.07
    Prov
    -0.07
    Wa
    -0.06
    救援
    -0.06
     ومن
    -0.06
    POSITIVE LOGITS
    妖
    0.08
    lya
    0.07
     Fisher
    0.07
    累积
    0.07
    ringe
    0.07
    컥
    0.07
    [color
    0.07
    *dt
    0.07
     jsx
    0.07
     SCALE
    0.07
    Act Density 0.002%

    No Known Activations