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

    Neuronpedia

    INDEX
    Explanations

    can find, see, customize, specify

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    إ
    1.11
     неожидан
    1.03
    I
    1.02
     держав
    1.02
    V
    1.02
    Воз
    1.01
    F
    1.01
    .
    0.98
    ة
    0.96
    <0x9C>
    0.96
    POSITIVE LOGITS
    но
    1.48
    ตรฐาน
    1.26
    ip
    1.20
     prostatic
    1.15
    sill
    1.15
    actment
    1.15
    sman
    1.14
    किला
    1.13
    𝐨
    1.13
    sizing
    1.12
    Act Density 0.374%

    No Known Activations