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

    Neuronpedia

    INDEX
    Explanations

    function calls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    }$$
    1.13
     baat
    1.13
    dlj
    1.12
     ℝ
    1.12
     afterthought
    1.09
    Ⓘ
    1.09
     ocho
    1.07
    }&
    1.05
    sai
    1.05
     milled
    1.04
    POSITIVE LOGITS
     yakin
    1.00
     массы
    0.97
    re
    0.96
     disorders
    0.95
     kiire
    0.91
     thách
    0.91
    瑩
    0.90
     seria
    0.90
    ন্যায়
    0.89
    因而
    0.89
    Act Density 0.018%

    No Known Activations