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

    Neuronpedia

    INDEX
    Explanations

    piecewise function

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ested
    -0.07
    .Release
    -0.06
     pz
    -0.06
    capacity
    -0.06
    ':
    -0.06
    Self
    -0.06
    Thing
    -0.06
     Kota
    -0.06
    Wonder
    -0.06
     lic
    -0.06
    POSITIVE LOGITS
    �
    0.07
    攻撃
    0.07
    EXPR
    0.07
    ['<{
    0.07
    _queryset
    0.06
    danger
    0.06
    ařilo
    0.06
    闪
    0.06
    .awtextra
    0.06
    loyd
    0.06
    Act Density 0.016%

    No Known Activations