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

    Neuronpedia

    INDEX
    Explanations

    fragment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     intrigue
    0.39
     hearing
    0.38
    走向
    0.38
     bureaus
    0.38
     microbe
    0.38
     juice
    0.37
     featured
    0.36
     buz
    0.36
     bubble
    0.36
     Gnome
    0.36
    POSITIVE LOGITS
     фраг
    0.46
    sockfd
    0.42
    fragment
    0.41
     fragmentary
    0.41
     centímetros
    0.41
     стек
    0.40
    sav
    0.39
     obtención
    0.39
     ущер
    0.38
    fragt
    0.38
    Act Density 0.000%

    No Known Activations