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

    Neuronpedia

    INDEX
    Explanations

    here:

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     MUST
    0.45
     înd
    0.43
     investir
    0.43
     sillonné
    0.42
    គ្ន
    0.41
     தயாரி
    0.41
    Must
    0.39
     informações
    0.39
     gestão
    0.39
     consag
    0.38
    POSITIVE LOGITS
    :
    0.48
    。
    0.45
    惑星
    0.42
    吗
    0.41
    мино
    0.41
    Reporter
    0.40
    建筑
    0.40
    度が
    0.39
    !=
    0.39
    ।
    0.39
    Act Density 0.032%

    No Known Activations