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

    Neuronpedia

    INDEX
    Explanations

    noun endings abstract nouns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ful
    1.01
    ning
    1.01
    estine
    0.95
    unky
    0.92
     concurring
    0.90
    々の
    0.90
    ah
    0.89
    ('{
    0.88
    ся
    0.87
    ningen
    0.86
    POSITIVE LOGITS
    ції
    1.16
    ции
    1.02
    ція
    1.01
    цию
    0.98
    asyon
    0.96
    েশনের
    0.96
    цію
    0.95
    ция
    0.95
    цій
    0.92
    aciju
    0.91
    Act Density 0.152%

    No Known Activations