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

    Neuronpedia

    INDEX
    Explanations

    curly braces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ers
    2.84
    ка
    2.72
     balas
    2.34
    \}$.
    2.26
    ENT
    2.25
    ❏
    2.18
    '%
    2.17
     foremost
    2.13
    inescent
    2.09
     travail
    2.05
    POSITIVE LOGITS
    Timber
    2.34
    要
    2.31
    り
    2.18
    eru
    2.12
    ोत्तम
    2.10
    er
    2.10
    Chứng
    2.07
    cropped
    2.04
    ㅂ
    2.03
    र
    2.01
    Act Density 0.060%

    No Known Activations