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

    Neuronpedia

    INDEX
    Explanations

    stating the goal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    begin
    1.19
    ↵
    1.11
     सारी
    1.09
     caro
    1.09
    rounded
    1.04
    TA
    1.02
    degrees
    1.00
    secrets
    0.99
    எம்
    0.98
    i
    0.97
    POSITIVE LOGITS
    ية
    1.21
    ير
    1.05
     $\|
    0.99
     aiming
    0.98
    анали
    0.98
     lucro
    0.97
    าะ
    0.97
    মাত্রা
    0.97
     foothold
    0.96
    ్
    0.96
    Act Density 0.186%

    No Known Activations