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

    Neuronpedia

    INDEX
    Explanations

    polymer analysis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    2.14
    i
    1.90
    d
    1.90
    an
    1.84
    os
    1.75
    as
    1.73
    は
    1.72
    ed
    1.70
    is
    1.67
    t
    1.63
    POSITIVE LOGITS
    ás
    1.05
    ल
    1.03
     offrent
    0.93
    슨
    0.92
     স্থানীয়
    0.90
    ્સ
    0.90
     effectuer
    0.90
    ы
    0.90
    ках
    0.89
     approche
    0.89
    Act Density 0.002%

    No Known Activations