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

    Neuronpedia

    INDEX
    Explanations

    top-notch, cutting-edge

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ların
    0.90
    AR
    0.87
    ﻤ
    0.85
    ને
    0.82
    Ẽ
    0.81
    ﻭ
    0.80
    $,
    0.80
    AT
    0.79
     براي
    0.75
    ดวก
    0.75
    POSITIVE LOGITS
    an
    0.95
    ik
    0.87
    ak
    0.79
    isasi
    0.79
    ;
    0.78
    ancing
    0.78
    izing
    0.77
    f
    0.77
    it
    0.73
    ンス
    0.70
    Act Density 0.026%

    No Known Activations