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

    Neuronpedia

    INDEX
    Explanations

    nuclear science and safety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Basta
    -0.82
     because
    -0.80
    Because
    -0.80
    ensan
    -0.77
    บ้าง
    -0.74
     tono
    -0.73
    ኑ
    -0.73
     antiquated
    -0.72
    thylene
    -0.71
    Centa
    -0.71
    POSITIVE LOGITS
    STRUCTION
    0.82
     jerk
    0.80
    ANAL
    0.80
     bà
    0.79
    бар
    0.79
    connected
    0.79
    classList
    0.79
    rás
    0.78
    ТО
    0.78
    daa
    0.78
    Act Density 0.023%

    No Known Activations