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

    Neuronpedia

    INDEX
    Explanations

    spanked, spied, spartan, spontaneously

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     grant
    0.90
    戞
    0.80
     grants
    0.75
     Grant
    0.75
    Grant
    0.73
     Grants
    0.71
    acije
    0.66
    grant
    0.64
     treads
    0.64
     آهن
    0.64
    POSITIVE LOGITS
     sp
    1.14
    sp
    0.98
     Sp
    0.88
    Sp
    0.81
    ats
    0.74
    小数
    0.73
     სპ
    0.71
     स्प
    0.70
    itten
    0.69
     Spar
    0.69
    Act Density 0.011%

    No Known Activations