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

    Neuronpedia

    INDEX
    Explanations

    as ... much, as ... long, as ... reported

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    т
    3.11
    ت
    3.09
    м
    3.03
    k
    2.50
    t
    2.41
    م
    2.39
    र
    2.17
    tay
    2.11
    त
    2.11
    л
    2.09
    POSITIVE LOGITS
    sembles
    2.59
    inine
    2.45
    ymmet
    2.03
     וכ
    1.95
    いたり
    1.84
    cribable
    1.78
    itada
    1.77
    ımın
    1.76
     vậy
    1.75
     כן
    1.73
    Act Density 0.210%

    No Known Activations