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

    Neuronpedia

    INDEX
    Explanations

    closing bracket followed by url

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     The
    0.58
    h
    0.57
    is
    0.53
     For
    0.52
    d
    0.51
     Y
    0.50
     This
    0.50
    S
    0.49
    R
    0.49
    l
    0.49
    POSITIVE LOGITS
     של
    0.69
     của
    0.65
     که
    0.63
     của
    0.60
    ۔
    0.60
     ११
    0.58
     måde
    0.58
    0.58
     šest
    0.56
    捒
    0.56
    Act Density 0.133%

    No Known Activations