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

    Neuronpedia

    INDEX
    Explanations

    Q4 quantization levels

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     customization
    0.43
     மற
    0.43
     customizations
    0.39
     Appear
    0.39
    thym
    0.39
    Bibli
    0.38
     Prepared
    0.38
     حف
    0.38
     хо
    0.37
     tông
    0.37
    POSITIVE LOGITS
     AFL
    0.54
    pley
    0.50
    נטי
    0.45
     NAB
    0.44
     nab
    0.40
    ваты
    0.40
     clubs
    0.39
    NAB
    0.39
     draft
    0.39
     naby
    0.38
    Act Density 0.003%

    No Known Activations