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

    Neuronpedia

    INDEX
    Explanations

    "tratta" / "trata" followed by "solo", "de", "un"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    恬
    0.64
    arbe
    0.62
    carotene
    0.61
    و
    0.59
    和他
    0.58
    咭
    0.58
    clust
    0.57
    herence
    0.57
    inthe
    0.57
    catalyzed
    0.57
    POSITIVE LOGITS
    ing
    0.68
    اد
    0.61
    í
    0.61
    т
    0.61
     adal
    0.58
     obvi
    0.58
     éve
    0.58
     SAV
    0.57
     overkill
    0.57
     repre
    0.56
    Act Density 0.009%

    No Known Activations