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

    Neuronpedia

    INDEX
    Explanations

    followed by description

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     그러니까
    0.79
     olmadığını
    0.75
    Итак
    0.75
    README
    0.70
    ():
    0.68
    actually
    0.68
     максимально
    0.66
    真正
    0.66
    (!)
    0.66
    underbrace
    0.64
    POSITIVE LOGITS
     ebenfalls
    1.79
     similar
    1.73
    こちらも
    1.62
     another
    1.55
     similaire
    1.52
     yine
    1.52
     again
    1.51
    これも
    1.50
     Similar
    1.50
    similar
    1.40
    Act Density 8.445%

    No Known Activations