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

    Neuronpedia

    INDEX
    Explanations

    code or function

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ط
    0.69
     (
    0.60
    त्या
    0.58
     "
    0.58
     همان
    0.57
     seconda
    0.56
    সুতরাং
    0.55
    Alternatively
    0.54
     أو
    0.54
     invariably
    0.54
    POSITIVE LOGITS
     કામ
    0.98
     stuff
    0.92
    ධාන
    0.90
     కృషి
    0.87
     stuffs
    0.86
    關鍵
    0.85
    关键
    0.82
     politics
    0.81
     Ingredients
    0.80
     работой
    0.79
    Act Density 0.291%

    No Known Activations