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

    Neuronpedia

    INDEX
    Explanations

    descriptive modifiers followed by nouns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ಎಲ್ಲಾ
    0.44
     усіх
    0.41
     endearing
    0.38
     식으로
    0.36
     እና
    0.35
     фаразлары
    0.35
     всіх
    0.35
     всех
    0.35
    )$
    0.34
     mọi
    0.34
    POSITIVE LOGITS
    e
    0.42
    ين
    0.42
    ka
    0.40
    Have
    0.40
    ي
    0.39
    6
    0.37
    2
    0.36
    5
    0.36
    7
    0.35
    Task
    0.35
    Act Density 0.551%

    No Known Activations