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

    Neuronpedia

    INDEX
    Explanations

    shuffling actions and randomization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     intracranial
    2.68
    不
    2.58
     aon
    2.48
     spese
    2.46
    rich
    2.43
    жды
    2.42
    rnd
    2.40
    atac
    2.39
    ners
    2.38
    rw
    2.37
    POSITIVE LOGITS
    tedir
    3.25
    ına
    2.78
     ढंग
    2.59
    ו
    2.59
    и
    2.54
     japonais
    2.53
    alım
    2.52
    tı
    2.49
    hus
    2.48
    تی
    2.46
    Act Density 0.036%

    No Known Activations