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

    Neuronpedia

    INDEX
    Explanations

    target

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     crec
    -0.08
    质
    -0.08
     simple
    -0.08
    Simple
    -0.07
     bool
    -0.07
     sederhana
    -0.07
     valid
    -0.07
     Building
    -0.07
     onderzoek
    -0.07
    Cong
    -0.07
    POSITIVE LOGITS
     unintended
    0.15
     inadvert
    0.11
     unint
    0.11
     نطاق
    0.11
     неж
    0.11
     inadvertently
    0.10
     specificity
    0.10
     undes
    0.10
     collateral
    0.10
     unwanted
    0.10
    Act Density 0.010%

    No Known Activations