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

    Neuronpedia

    INDEX
    Explanations

    for key, value

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     presenceData
    0.41
     আজি
    0.38
    讵
    0.38
    authState
    0.38
     hasattr
    0.35
     Streit
    0.35
    逖
    0.35
    тельной
    0.35
     validInput
    0.35
    ەت
    0.35
    POSITIVE LOGITS
     key
    1.40
    key
    1.22
     keys
    1.05
     k
    1.02
     клю
    1.00
     ключ
    0.99
     ключе
    0.97
    鍵
    0.96
    キー
    0.96
     Key
    0.93
    Act Density 0.014%

    No Known Activations