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

    Neuronpedia

    INDEX
    Explanations

    `ade` followed by `cre` or `class`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    un
    0.93
    as
    0.90
    o
    0.88
    il
    0.79
    ul
    0.79
    n
    0.76
    um
    0.75
    not
    0.72
    ate
    0.68
    ov
    0.67
    POSITIVE LOGITS
    ורי
    0.69
    ков
    0.68
     outstretched
    0.65
     evaded
    0.63
    ні
    0.63
    ким
    0.61
    קד
    0.60
    ковского
    0.59
    شي
    0.59
    кин
    0.59
    Act Density 0.000%

    No Known Activations