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

    Neuronpedia

    INDEX
    Explanations

    working with numbers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     manifest
    -0.07
    Most
    -0.07
     úřad
    -0.07
    tracts
    -0.06
     Snapchat
    -0.06
    -tracking
    -0.06
    ographers
    -0.06
    $x
    -0.06
    _lo
    -0.06
     Calories
    -0.06
    POSITIVE LOGITS
     "'.
    0.07
    SORT
    0.07
    .ref
    0.06
     Bü
    0.06
    estring
    0.06
     توان
    0.06
     taxing
    0.06
    �
    0.06
     reluct
    0.06
    _pg
    0.06
    Act Density 0.099%

    No Known Activations