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

    Neuronpedia

    INDEX
    Explanations

    test case IDs or document IDs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    relative
    0.36
    does
    0.35
    アウト
    0.35
    瀾
    0.35
     Mixture
    0.34
     জুড়ে
    0.34
    ocardial
    0.34
    hampton
    0.34
    🅘
    0.34
     fre
    0.33
    POSITIVE LOGITS
    ्वे
    0.41
    люми
    0.41
    Tipo
    0.40
    Seed
    0.39
    Stef
    0.39
     венти
    0.39
    Cancer
    0.38
    Ла
    0.38
    textFile
    0.38
    ኼ
    0.38
    Act Density 0.002%

    No Known Activations