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

    Neuronpedia

    INDEX
    Explanations

    proximity and reachability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     longer
    0.46
    lama
    0.40
     lama
    0.40
     beyond
    0.40
     넓
    0.39
    وسیع
    0.39
     benötigen
    0.39
     längre
    0.39
     uzun
    0.39
     साढ़े
    0.37
    POSITIVE LOGITS
     vicinity
    0.56
     reasonable
    0.48
     dễ
    0.48
     mudah
    0.47
     proximité
    0.45
     proximity
    0.45
     разум
    0.43
     fácil
    0.43
     услы
    0.40
     близь
    0.40
    Act Density 0.015%

    No Known Activations