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

    Neuronpedia

    INDEX
    Explanations

    multilingual script identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     a
    1.67
    न
    1.57
    ン
    1.53
     at
    1.45
    한
    1.35
    个
    1.21
    ą
    1.20
    ة
    1.16
    ม
    1.13
    К
    1.10
    POSITIVE LOGITS
    uje
    1.34
    c
    1.15
    ן
    1.14
    ז
    1.05
    p
    1.03
    িন
    1.02
    ет
    1.02
    g
    1.00
    ih
    0.97
    she
    0.96
    Act Density 0.349%

    No Known Activations