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

    Neuronpedia

    INDEX
    Explanations

    pervasive influence and impact

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    c
    0.73
    트
    0.70
    리
    0.66
    re
    0.63
    é
    0.63
     prompts
    0.59
    r
    0.59
    ير
    0.58
    の間
    0.57
    EZ
    0.56
    POSITIVE LOGITS
    ing
    0.78
    ة
    0.75
    0
    0.70
    <0x80>
    0.65
    al
    0.63
    filed
    0.61
    frac
    0.59
     лиш
    0.58
    ૦
    0.58
    دود
    0.56
    Act Density 0.004%

    No Known Activations