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

    Neuronpedia

    INDEX
    Explanations

    getting worse or fading

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ak
    1.48
    годи
    1.41
    na
    1.40
     وعند
    1.40
    ka
    1.39
    ка
    1.37
    ma
    1.36
    িক
    1.36
    ве
    1.29
    नी
    1.28
    POSITIVE LOGITS
    '`--
    1.23
    ᕝ
    1.23
    slideDuplicate
    1.20
    ᓗ
    1.19
    )}_{
    1.18
    )}(\
    1.17
    )},
    1.16
    )})
    1.16
    遐
    1.16
    udp
    1.15
    Act Density 0.224%

    No Known Activations