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

    Neuronpedia

    INDEX
    Explanations

    stripping whitespace

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     steeply
    1.26
    े
    1.25
    ları
    1.23
    ד
    1.23
     acreage
    1.23
    হ
    1.22
     age
    1.19
    schaft
    1.19
    taş
    1.16
     olay
    1.15
    POSITIVE LOGITS
    Вот
    1.46
    قیع
    1.39
    𒋗
    1.38
     используется
    1.36
    鳅
    1.35
    靼
    1.30
    汋
    1.29
    ToReference
    1.28
    <unused279>
    1.27
     Tibetan
    1.25
    Act Density 0.001%

    No Known Activations