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

    Neuronpedia

    INDEX
    Explanations

    these or all

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    有两种
    0.94
     данного
    0.91
     данной
    0.89
     consiste
    0.82
    টিতে
    0.82
    this
    0.74
    在该
    0.72
     denna
    0.71
     означает
    0.71
     signifie
    0.71
    POSITIVE LOGITS
    這些
    2.42
    这些
    2.42
    These
    2.40
     these
    2.40
     These
    2.27
     这些
    2.26
    these
    2.19
     all
    2.17
    これらの
    2.16
     semuanya
    2.06
    Act Density 0.823%

    No Known Activations