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

    Neuronpedia

    INDEX
    Explanations

    lots of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     at
    1.16
     of
    1.15
    the
    1.14
     was
    1.08
    of
    1.05
                
    1.04
    The
    1.01
    t
    1.00
    0
    0.99
     d
    0.98
    POSITIVE LOGITS
    чи
    1.12
    ர
    0.98
    ަ
    0.94
    ст
    0.90
    ли
    0.90
     avanzar
    0.89
    бто
    0.89
     igualmente
    0.88
    нди
    0.88
     ocasion
    0.87
    Act Density 0.004%

    No Known Activations