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

    Neuronpedia

    INDEX
    Explanations

    is/are followed by descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    情報は
    0.67
     היא
    0.66
     lässt
    0.64
    গুলোর
    0.62
    こちらは
    0.58
    ផ្សេង
    0.57
     is
    0.56
    等は
    0.56
    들은
    0.56
    これは
    0.55
    POSITIVE LOGITS
     very
    1.43
     molto
    1.43
     muy
    1.35
     매우
    1.31
     foarte
    1.30
     velmi
    1.30
     très
    1.28
     bardzo
    1.28
     sehr
    1.23
     очень
    1.23
    Act Density 0.002%

    No Known Activations