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

    Neuronpedia

    INDEX
    Explanations

    command preceded by word or token

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    is
    0.89
    h
    0.87
    j
    0.82
    yb
    0.80
    sale
    0.77
    ד
    0.77
    g
    0.73
    hh
    0.73
    hg
    0.72
    hhhh
    0.71
    POSITIVE LOGITS
    Қ
    1.05
    ে
    1.02
    ε
    1.00
    いますが
    0.97
    Cadastro
    0.95
    ી
    0.95
    混凝土
    0.94
    நிலை
    0.93
     EnglishChoose
    0.93
    Muscle
    0.93
    Act Density 0.004%

    No Known Activations