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

    Neuronpedia

    INDEX
    Explanations

    breakdown with explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ၂
    0.54
    _
    0.48
     vaan
    0.47
     eerst
    0.47
     peux
    0.45
    fromj
    0.45
     numai
    0.45
     ၁
    0.45
     නමුත්
    0.45
     GmbH
    0.44
    POSITIVE LOGITS
    s
    0.72
    ی
    0.62
    ات
    0.61
     to
    0.60
    r
    0.60
    re
    0.59
    其他
    0.58
    то
    0.57
    t
    0.57
    야
    0.57
    Act Density 0.032%

    No Known Activations