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

    Neuronpedia

    INDEX
    Explanations

    once you paste

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     remembering
    0.93
     aceasta
    0.90
     this
    0.90
     tämä
    0.89
     personalization
    0.84
     바랍니다
    0.83
     anybody
    0.82
     Remembering
    0.82
     opcion
    0.80
     これ
    0.80
    POSITIVE LOGITS
    Once
    1.52
     Once
    1.50
     once
    1.47
    once
    1.34
    一旦
    0.89
     एकदा
    0.82
     sekali
    0.77
     eens
    0.77
     fois
    0.68
    一度
    0.67
    Act Density 0.144%

    No Known Activations