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

    Neuronpedia

    INDEX
    Explanations

    Hollywood sign

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     трябва
    0.61
    阍
    0.60
     është
    0.59
     должна
    0.59
     πρέπει
    0.58
    '
    0.57
    ﾗ
    0.57
     تكون
    0.57
     ẞ
    0.57
    𝓈
    0.56
    POSITIVE LOGITS
    us
    0.94
    ul
    0.61
    ol
    0.60
    un
    0.59
    in
    0.58
    ahari
    0.58
    ah
    0.56
    az
    0.56
    eng
    0.55
    t
    0.55
    Act Density 0.001%

    No Known Activations