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

    Neuronpedia

    INDEX
    Explanations

    complex topics

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (progress
    -0.07
    社会
    -0.07
    _zip
    -0.07
     Julian
    -0.07
     Trans
    -0.07
     Lun
    -0.07
     naz
    -0.07
     khi
    -0.07
     corresponds
    -0.07
     san
    -0.06
    POSITIVE LOGITS
     مليون
    0.07
    幄
    0.07
    -taking
    0.07
     newData
    0.07
    沥
    0.07
    艰
    0.06
    _geometry
    0.06
    בדק
    0.06
    stats
    0.06
    _area
    0.06
    Act Density 0.387%

    No Known Activations