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

    Neuronpedia

    INDEX
    Explanations

    Okay, starting response

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     people
    0.58
     it
    0.55
     time
    0.53
     he
    0.52
     at
    0.51
     we
    0.51
     good
    0.50
     you
    0.50
     there
    0.50
     so
    0.49
    POSITIVE LOGITS
    Описание
    0.53
     Мол
    0.53
    Ви
    0.52
    édition
    0.52
    genCode
    0.52
    概述
    0.51
    Conceptual
    0.51
     объем
    0.51
    Editor
    0.50
     સંપૂર્ણ
    0.50
    Act Density 4.143%

    No Known Activations