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

    Neuronpedia

    INDEX
    Explanations

    at a level or scale

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    这一切
    0.19
     এটাই
    0.17
    UIApplication
    0.17
    $_{
    0.17
     পরের
    0.17
    benefits
    0.17
    这项
    0.17
    $-[
    0.17
    cancer
    0.17
     требованиям
    0.17
    POSITIVE LOGITS
     scale
    0.24
     runtime
    0.24
     glance
    0.23
     various
    0.23
    mes
    0.23
     timescales
    0.22
     constrained
    0.21
     locations
    0.21
     interfaces
    0.21
     grit
    0.21
    Act Density 0.069%

    No Known Activations