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

    Neuronpedia

    INDEX
    Explanations

    why things work or are important

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ありますが
    0.32
    ımı
    0.32
    ду
    0.31
    {-
    0.31
    !'
    0.31
    க்
    0.30
    !
    0.30
    ವುದು
    0.30
     सहित
    0.29
    ς
    0.29
    POSITIVE LOGITS
    in
    0.49
    на
    0.37
     de
    0.36
    ת
    0.35
    ين
    0.35
    at
    0.35
    ే
    0.34
    त
    0.34
    𒄑
    0.34
     on
    0.32
    Act Density 0.544%

    No Known Activations