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

    Neuronpedia

    INDEX
    Explanations

    say "use"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .Pin
    -0.08
     demonstration
    -0.07
    高铁
    -0.07
    Limited
    -0.07
    .POS
    -0.07
    .int
    -0.07
    ::-
    -0.07
    /her
    -0.07
    SupportedException
    -0.06
     hom
    -0.06
    POSITIVE LOGITS
     blade
    0.07
     keyboards
    0.07
     ankle
    0.07
    ächst
    0.07
    Weapon
    0.07
     применя
    0.07
    mounted
    0.06
    _ARROW
    0.06
    wództw
    0.06
     Edinburgh
    0.06
    Act Density 0.003%

    No Known Activations