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

    Neuronpedia

    INDEX
    Explanations

    tuples and pairs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     todos
    0.45
     todas
    0.44
    ताज
    0.41
     aloe
    0.41
    Todas
    0.41
     toate
    0.40
     všechny
    0.40
     Elektrokhimiya
    0.40
     všetky
    0.39
     resultContent
    0.39
    POSITIVE LOGITS
     tuple
    1.27
     pair
    1.17
    pair
    1.16
     pairs
    1.12
     duo
    1.09
    tuple
    1.05
    pairs
    1.04
     tuples
    1.02
    Tuple
    1.01
     Tuple
    0.96
    Act Density 0.628%

    No Known Activations