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

    Neuronpedia

    INDEX
    Explanations

    counting items, cards, gases, and years

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     snippet
    0.70
     carattere
    0.69
     слой
    0.68
     concetto
    0.66
     scenario
    0.66
     systeem
    0.66
     Grenze
    0.65
    这个人
    0.65
     perencanaan
    0.65
     metodologia
    0.64
    POSITIVE LOGITS
     items
    1.16
     files
    1.11
     cells
    1.10
     songs
    1.07
     rows
    1.06
     papers
    1.05
     courses
    1.03
     countries
    1.01
     books
    0.97
     objects
    0.96
    Act Density 2.884%

    No Known Activations