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

    Neuronpedia

    INDEX
    Explanations

    subscript 2 and iosis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    0.98
    5
    0.87
    ;
    0.84
    ’
    0.83
    $
    0.72
    ה
    0.70
    S
    0.69
    !
    0.69
    ه
    0.67
    ↵↵
    0.65
    POSITIVE LOGITS
    u
    0.86
     уены
    0.74
    ㏄
    0.73
     дальнейшем
    0.72
    밂
    0.71
    GBuf
    0.69
    jhelp
    0.69
     pertinentes
    0.68
    抽選
    0.68
    uoj
    0.67
    Act Density 0.001%

    No Known Activations