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

    Neuronpedia

    INDEX
    Explanations

    behind the scenes, stage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    at
    1.55
    and
    1.45
    as
    1.40
    1.35
     (
    1.24
    am
    1.22
     Artillery
    1.21
    et
    1.20
    ти
    1.19
    en
    1.17
    POSITIVE LOGITS
    ями
    1.41
    𝐣
    1.37
    ቝ
    1.34
     emphasise
    1.33
    дати
    1.30
    𝐪
    1.30
     conexiones
    1.30
    𒅆
    1.27
    뇰
    1.27
    гээ
    1.26
    Act Density 0.038%

    No Known Activations