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

    Neuronpedia

    INDEX
    Explanations

    section headers and labels

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Bạn
    0.29
    Jeśli
    0.29
    它们
    0.27
    Как
    0.26
    如果你
    0.26
    हर
    0.25
    Après
    0.25
    if
    0.25
    Shroud
    0.25
    आप
    0.25
    POSITIVE LOGITS
    ing
    0.35
    টি
    0.34
     from
    0.34
     consists
    0.34
     strength
    0.33
     tersebut
    0.32
     starts
    0.31
     found
    0.31
     started
    0.31
     consisting
    0.31
    Act Density 0.417%

    No Known Activations