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

    Neuronpedia

    INDEX
    Explanations

    backoff and vasodilation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    れ
    0.82
     Labrenzia
    0.77
    vođa
    0.77
    ură
    0.77
    histoire
    0.76
    Ż
    0.76
    af
    0.75
    hah
    0.75
    хие
    0.73
    í
    0.73
    POSITIVE LOGITS
    ↵↵
    1.18
    ك
    1.13
     on
    1.02
     ل
    0.89
    ↵
    0.85
     for
    0.85
    、
    0.85
     ส
    0.83
    ،
    0.83
     rápido
    0.83
    Act Density 0.000%

    No Known Activations