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

    Neuronpedia

    INDEX
    Explanations

    using for and then descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     scum
    0.45
     Buf
    0.43
    тена
    0.43
     Src
    0.42
     После
    0.41
     courage
    0.39
     Пара
    0.39
    цкий
    0.39
     Ook
    0.39
    ೇವೆ
    0.39
    POSITIVE LOGITS
    dispers
    0.47
     disperse
    0.46
    बो
    0.46
     πολλ
    0.45
    ｢
    0.44
     dispersal
    0.44
     unsold
    0.43
    Lamb
    0.43
     algod
    0.42
    贬
    0.42
    Act Density 0.002%

    No Known Activations