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

    Neuronpedia

    INDEX
    Explanations

    word beginnings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    т
    2.55
     inférieures
    1.66
    ння
    1.65
     latérales
    1.65
    𝓻
    1.62
     rétrécies
    1.61
    σ
    1.60
    客様
    1.59
     хорошее
    1.59
    та
    1.53
    POSITIVE LOGITS
    ار
    1.93
    습니다
    1.81
    soever
    1.76
    arası
    1.75
    िक
    1.73
    িল
    1.70
    ä
    1.66
     underestimate
    1.64
    as
    1.62
    ﮈ
    1.62
    Act Density 0.174%

    No Known Activations