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

    Neuronpedia

    INDEX
    Explanations

    Chinatown, populist, keV, onclick

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    "
    -3.41
    ↵
    -3.14
     It
    -2.81
    ’
    -2.77
    на
    -2.73
     Not
    -2.73
    d
    -2.63
    l
    -2.55
    section
    -2.48
    ?
    -2.38
    POSITIVE LOGITS
    鯔
    3.17
    他的
    2.92
    샅
    2.91
    ㍑
    2.80
    皤
    2.78
    ╽
    2.75
    2.72
    樒
    2.69
    튿
    2.67
    geſ
    2.64
    Act Density 0.012%

    No Known Activations