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

    Neuronpedia

    INDEX
    Explanations

    inaccuracy, limitations, exaggeration

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    υ
    0.65
    ı
    0.50
    ui
    0.48
    の変化
    0.48
    น
    0.48
    ン
    0.46
    の
    0.46
    ικούς
    0.42
    のマ
    0.42
    ری
    0.42
    POSITIVE LOGITS
    D
    0.65
    H
    0.59
    L
    0.58
    0.56
    Y
    0.50
    .
    0.50
    P
    0.49
    W
    0.48
    J
    0.48
    ↵↵
    0.48
    Act Density 0.210%

    No Known Activations