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

    Neuronpedia

    INDEX
    Explanations

    list item topics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mafia
    1.05
     hairdresser
    0.98
     stumps
    0.91
     smells
    0.90
    哪怕
    0.90
     underwear
    0.90
     bici
    0.89
    movieId
    0.89
     vampires
    0.89
     yeast
    0.88
    POSITIVE LOGITS
    Improved
    0.83
     Improved
    0.82
     Reduced
    0.80
     improved
    0.80
    ·
    0.78
     tasarım
    0.76
     Technological
    0.76
     উন্নত
    0.75
     Flexible
    0.75
     ·
    0.73
    Act Density 0.758%

    No Known Activations