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

    Neuronpedia

    INDEX
    Explanations

    remains unchanged / continues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    도
    1.89
    是为了
    1.88
    是从
    1.80
    是可以
    1.75
    EYER
    1.75
    ס
    1.75
    ECTOR
    1.74
    是最
    1.71
    ⎮
    1.71
     scour
    1.69
    POSITIVE LOGITS
    p
    2.61
    t
    2.50
    b
    2.48
    n
    2.28
    på
    2.23
     रहकर
    2.08
    l
    2.03
    j
    1.95
    un
    1.95
    uk
    1.91
    Act Density 0.776%

    No Known Activations