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

    Neuronpedia

    INDEX
    Explanations

    whispered secrets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    茈
    0.52
     будут
    0.52
    यर
    0.51
    способ
    0.50
    策略
    0.48
    潾
    0.48
    어가
    0.47
    还
    0.47
     ersch
    0.47
    修复
    0.47
    POSITIVE LOGITS
    people
    0.51
    states
    0.46
    io
    0.46
    glue
    0.45
    pre
    0.45
    France
    0.45
    stable
    0.44
    Glue
    0.44
    bigcirc
    0.44
    ew
    0.43
    Act Density 0.000%

    No Known Activations