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

    Neuronpedia

    INDEX
    Explanations

    concrete nouns

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     અન્ય
    0.62
     Auxiliary
    0.60
     Community
    0.54
     local
    0.54
     Archive
    0.53
     Former
    0.52
     discretionary
    0.52
     Journals
    0.52
     വ്യക്തമാക്കി
    0.52
     Private
    0.51
    POSITIVE LOGITS
     broccoli
    1.17
     sailboat
    1.14
     ketchup
    1.14
     hamburgers
    1.14
     airplane
    1.11
     pizza
    1.10
     skyscrapers
    1.10
     raindrops
    1.09
     skyscraper
    1.08
     basketball
    1.06
    Act Density 1.830%

    No Known Activations