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

    Neuronpedia

    INDEX
    Explanations

    от a

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     desp
    0.82
     enhancement
    0.77
     senza
    0.77
     aper
    0.77
     sans
    0.75
     carousel
    0.75
     zu
    0.73
     appe
    0.73
     inf
    0.72
     suspension
    0.71
    POSITIVE LOGITS
    گ
    1.00
    gr
    0.99
    sp
    0.98
    br
    0.92
    су
    0.92
    nos
    0.92
    ле
    0.91
    сы
    0.90
    b
    0.90
    born
    0.89
    Act Density 0.030%

    No Known Activations