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

    Neuronpedia

    INDEX
    Explanations

    pick a theme

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    (**
    0.47
    聪明
    0.45
    enee
    0.41
    ücksicht
    0.40
     communicator
    0.40
    (*)
    0.38
    (&
    0.38
     intelligible
    0.38
    ('
    0.38
    諳
    0.38
    POSITIVE LOGITS
     Alvar
    0.53
     tema
    0.51
    テーマ
    0.50
    Tema
    0.47
     Thema
    0.46
     Solve
    0.46
     книги
    0.46
     Same
    0.45
    問題を
    0.45
     Themen
    0.45
    Act Density 0.000%

    No Known Activations