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

    Neuronpedia

    INDEX
    Explanations

    discount, 4-bit, Tensor, img, df

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    と思いますが
    0.50
    !”,
    0.45
    かもしれませんが
    0.44
    ですし
    0.44
     …,
    0.44
    !!!");
    0.44
    ****",
    0.44
    】【
    0.43
     isang
    0.43
     없고
    0.43
    POSITIVE LOGITS
    ）
    0.63
    )
    0.52
    。
    0.45
    .
    0.44
    ؟
    0.43
    ។
    0.42
    :)
    0.41
    Ꭲ
    0.41
    」
    0.41
    ‍
    0.40
    Act Density 0.501%

    No Known Activations