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

    Neuronpedia

    INDEX
    Explanations

    improved efficiency and demand

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    dg
    0.44
     monarchy
    0.43
    ứa
    0.43
     Ancient
    0.42
     Ngày
    0.41
     Sq
    0.41
    الحمد
    0.40
    broke
    0.40
    amil
    0.39
    centre
    0.39
    POSITIVE LOGITS
    ガイド
    0.46
    と同様
    0.42
    ัติ
    0.41
     メ
    0.40
     マルチ
    0.39
     に
    0.39
     방식
    0.39
     バス
    0.39
    調査
    0.39
     DENUMIRE
    0.39
    Act Density 0.001%

    No Known Activations