INDEX
    Explanations

    This neuron fires on the initial “L” at the start of each post’s title/header line.

    New Auto-Interp
    Negative Logits
     arz
    -0.07
     Unix
    -0.06
     SEX
    -0.06
     Yield
    -0.06
    фт
    -0.06
     Rank
    -0.06
     Penguin
    -0.06
     colorful
    -0.06
     yield
    -0.06
     Script
    -0.06
    POSITIVE LOGITS
    ıldığı
    0.07
    -md
    0.07
     imperative
    0.07
     благодаря
    0.06
     vu
    0.06
    _Instance
    0.06
    maması
    0.06
     мы
    0.06
    ократи
    0.06
     принадлеж
    0.06
    Act Density 0.004%

    No Known Activations