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

    Neuronpedia

    INDEX
    Explanations

    prices and percentages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     software
    0.45
     receptive
    0.44
     orchestr
    0.41
     brist
    0.40
     based
    0.39
     humor
    0.39
     sequential
    0.38
     using
    0.37
     behavior
    0.37
     얼
    0.37
    POSITIVE LOGITS
    three
    0.77
    two
    0.71
    five
    0.70
     trzech
    0.67
     două
    0.67
     двумя
    0.66
    3
    0.64
     vijf
    0.64
     ढाई
    0.64
     два
    0.63
    Act Density 0.782%

    No Known Activations