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

    Neuronpedia

    INDEX
    Explanations

    streets

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     GraphQL
    -0.07
     jump
    -0.07
    adb
    -0.07
    _basis
    -0.07
    adar
    -0.07
    dance
    -0.07
    cdf
    -0.06
    Fold
    -0.06
    ornado
    -0.06
    _sq
    -0.06
    POSITIVE LOGITS
     streets
    0.10
     Mattis
    0.07
     Streets
    0.07
    [path
    0.06
     Heroes
    0.06
     roads
    0.06
    .bs
    0.06
    �
    0.06
    تغ
    0.06
     jclass
    0.06
    Act Density 0.005%

    No Known Activations