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

    Neuronpedia

    INDEX
    Explanations

    making things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Uppercase
    1.25
    Appro
    1.21
    hobby
    1.19
    enade
    1.17
    spaper
    1.17
    bpy
    1.17
    Guides
    1.16
    commuting
    1.15
     auditorium
    1.15
    Founded
    1.12
    POSITIVE LOGITS
    いますが
    1.14
     জন্যই
    1.13
     esa
    1.11
     juga
    1.08
     частности
    1.05
     sair
    1.01
     tentar
    0.99
     meisten
    0.97
    ﺎ
    0.96
     vers
    0.96
    Act Density 0.004%

    No Known Activations