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

    Neuronpedia

    INDEX
    Explanations

    binary search

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    un
    0.58
    ul
    0.57
    failed
    0.56
    ac
    0.55
    arc
    0.53
    um
    0.52
    lines
    0.51
    id
    0.50
    os
    0.50
    il
    0.49
    POSITIVE LOGITS
     alguna
    0.49
     democracia
    0.48
     aqu
    0.48
    銭
    0.47
     вид
    0.47
     aprob
    0.44
     obten
    0.44
     banheiro
    0.44
     democracy
    0.43
     combina
    0.43
    Act Density 0.006%

    No Known Activations