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

    Neuronpedia

    INDEX
    Explanations

    collapsible lists or IP/PDF attacks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ⃣
    1.80
    ️⃣
    1.71
     underlies
    1.53
    '
    1.47
    يه
    1.45
    ่อง
    1.42
    的工作
    1.41
    та
    1.37
     gewährle
    1.37
    的东西
    1.35
    POSITIVE LOGITS
    src
    1.23
    ၄
    1.21
    washer
    1.20
    ن
    1.19
    K
    1.17
    你
    1.17
    ၃
    1.16
    KIT
    1.16
    wehr
    1.14
    JPanel
    1.14
    Act Density 0.001%

    No Known Activations