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

    Neuronpedia

    INDEX
    Explanations

    code `id` or `find`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     volver
    1.02
     vivió
    0.96
     ahorrar
    0.95
     ଦ
    0.92
     ଗ
    0.92
    摶
    0.91
     ସ
    0.90
     producir
    0.89
     más
    0.89
     aceptar
    0.88
    POSITIVE LOGITS
    ENO
    0.73
    NES
    0.66
    SF
    0.65
    бари
    0.64
    خ
    0.63
     гости
    0.62
    ESTIC
    0.62
    elines
    0.61
    RED
    0.61
    WIRE
    0.61
    Act Density 0.001%

    No Known Activations