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

    Neuronpedia

    INDEX
    Explanations

    for clarity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    in
    0.36
    y
    0.34
    ed
    0.29
    e
    0.27
    o
    0.27
    es
    0.26
    em
    0.26
    ()
    0.25
    ar
    0.25
    as
    0.25
    POSITIVE LOGITS
     sake
    0.26
    ked
    0.24
    казыва
    0.22
     profissionais
    0.21
     приобре
    0.21
     satisfacer
    0.21
     purposes
    0.21
     поддержа
    0.20
     estudiantes
    0.20
     personas
    0.20
    Act Density 0.638%

    No Known Activations