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

    Neuronpedia

    INDEX
    Explanations

    multi-script characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    de
    1.33
    te
    1.26
    ri
    1.24
    ry
    1.17
    sa
    1.17
    1.14
    were
    1.13
    re
    1.11
    s
    1.11
    ra
    1.10
    POSITIVE LOGITS
    kých
    1.43
    ر
    1.36
    り
    1.35
    aków
    1.30
    在
    1.30
    ون
    1.26
     في
    1.18
    L
    1.18
    <0xD5>
    1.17
    }"
    1.16
    Act Density 0.000%

    No Known Activations