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

    Neuronpedia

    INDEX
    Explanations

    research purpose, paper, proposal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    1.32
    م
    1.23
    การ
    1.20
    心
    0.95
    问题
    0.93
    信息
    0.93
    ح
    0.93
    ки
    0.92
    位置
    0.91
    工作
    0.89
    POSITIVE LOGITS
    a
    1.20
    ט
    1.16
     can
    1.13
    सँग
    1.06
    ;
    1.05
    К
    1.02
    n
    1.01
    nD
    1.00
    in
    0.99
    al
    0.99
    Act Density 0.031%

    No Known Activations