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

    Neuronpedia

    INDEX
    Explanations

    Tokyo skyscrapers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cottage
    0.45
    炕
    0.45
    鹌
    0.45
     dissociated
    0.43
     Hardware
    0.43
     sawdust
    0.42
     Monteiro
    0.42
     Disqus
    0.42
     cottage
    0.42
    Subgraph
    0.42
    POSITIVE LOGITS
     skyscraper
    2.02
     skyscrapers
    1.88
     towers
    1.73
     tower
    1.73
    skyscraper
    1.52
    Tower
    1.48
     towering
    1.48
     Skyscrapers
    1.46
     tallest
    1.45
    tower
    1.45
    Act Density 0.027%

    No Known Activations