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

    Neuronpedia

    INDEX
    Explanations

    one of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     związ
    2.62
    læ
    2.08
    preco
    2.03
    事务
    2.02
    2.00
    deficient
    1.99
    dont
    1.97
    hetical
    1.94
    phenylsulfanyl
    1.94
    nière
    1.92
    POSITIVE LOGITS
    р
    3.10
    ی
    2.77
    र
    2.28
    κ
    2.28
     heck
    2.01
    씩
    2.00
    ल
    1.93
    жды
    1.91
    ша
    1.88
    ри
    1.86
    Act Density 0.028%

    No Known Activations