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

    Neuronpedia

    INDEX
    Explanations

    declaring types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    全球
    0.49
     douleur
    0.43
    VENUE
    0.43
    RENDITION
    0.42
    হো
    0.42
     رسول
    0.41
    特典
    0.40
    脱毛
    0.40
    écou
    0.40
    Stewart
    0.39
    POSITIVE LOGITS
     type
    1.27
    类型
    1.23
     类型
    1.16
    type
    1.11
     타입을
    1.10
     Type
    1.07
     types
    1.06
     тип
    1.06
     타입
    1.06
    Type
    1.05
    Act Density 0.053%

    No Known Activations