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

    Neuronpedia

    INDEX
    Explanations

    strength, weakness, and power

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    cosh
    0.87
     গুলি
    0.83
     Space
    0.82
    zor
    0.81
     spazi
    0.78
    োন্ন
    0.77
     espaces
    0.76
     eeu
    0.75
    িজ
    0.75
     espacios
    0.75
    POSITIVE LOGITS
    hold
    1.14
    holds
    1.10
    িশালী
    1.09
     mẽ
    1.01
     💪
    0.99
     weak
    0.91
    Weak
    0.91
    weak
    0.90
     winds
    0.89
    strong
    0.87
    Act Density 0.499%

    No Known Activations