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

    Neuronpedia

    INDEX
    Explanations

    really, real, internal, watertight

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,!
    0.57
    \".
    0.57
    nict
    0.55
     Implementation
    0.52
     Awesome
    0.51
    ;.
    0.51
     designates
    0.51
     will
    0.50
    חת
    0.50
     Includes
    0.50
    POSITIVE LOGITS
    ia
    0.65
    ی
    0.64
    یوں
    0.59
    ियों
    0.55
    ị
    0.55
    </code>
    0.54
    ٹ
    0.54
    یس
    0.53
    am
    0.53
    itements
    0.52
    Act Density 0.000%

    No Known Activations