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

    Neuronpedia

    INDEX
    Explanations

    hash function and int

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     involves
    0.55
     from
    0.46
    Ladder
    0.46
     escal
    0.45
     ladder
    0.45
     involved
    0.44
     ground
    0.43
     around
    0.43
     at
    0.42
     invoked
    0.42
    POSITIVE LOGITS
     tentar
    0.52
    籵
    0.52
    xlink
    0.51
     Tonic
    0.49
    avacan
    0.48
    ptitle
    0.48
    Tovar
    0.48
    arq
    0.48
     satisfait
    0.47
     Дет
    0.47
    Act Density 0.000%

    No Known Activations