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

    Neuronpedia

    INDEX
    Explanations

    tuples, pairs, and triplets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Brands
    0.75
    brands
    0.72
    brand
    0.71
     brands
    0.67
     gate
    0.65
     brand
    0.64
    lop
    0.62
    Companies
    0.62
     dall
    0.61
    ிகா
    0.60
    POSITIVE LOGITS
     tuples
    2.12
     tuple
    2.09
    tuple
    1.94
    Tuple
    1.91
     Tuple
    1.90
    tuples
    1.77
     pairs
    1.71
     Pairs
    1.65
    pairs
    1.57
     pair
    1.57
    Act Density 0.436%

    No Known Activations