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

    Neuronpedia

    INDEX
    Explanations

    Dune, Linguistic, Distant, Socioeconomic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    0.39
    y
    0.38
    e
    0.37
     (
    0.36
    s
    0.36
     input
    0.35
    ,
    0.35
     [
    0.33
     in
    0.33
    es
    0.32
    POSITIVE LOGITS
     ाट
    0.45
     gobierno
    0.45
    𒅗
    0.43
    trashItem
    0.40
    <unused2134>
    0.40
    ibrant
    0.40
     চাহিয়
    0.40
    嵃
    0.40
     britannique
    0.40
     rije
    0.39
    Act Density 0.335%

    No Known Activations