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

    Neuronpedia

    INDEX
    Explanations

    observe or find something

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    re
    -1.74
    x
    -1.69
    <
    -1.48
    y
    -1.48
    ri
    -1.48
    𝐋
    -1.47
    ch
    -1.46
    nt
    -1.45
    g
    -1.41
    k
    -1.41
    POSITIVE LOGITS
     recibió
    1.32
    это
    1.30
    スズキ
    1.30
    nó
    1.30
     detención
    1.30
     dão
    1.29
     comenzó
    1.29
     vão
    1.27
    日産
    1.26
     comienza
    1.25
    Act Density 0.067%

    No Known Activations