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

    Neuronpedia

    INDEX
    Explanations

    at the beginning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    敚
    0.39
     Poul
    0.37
     формування
    0.36
    icuous
    0.35
    驽
    0.35
    িমের
    0.35
     Harrier
    0.34
    ということです
    0.33
    াইজ
    0.33
     మాత్రం
    0.33
    POSITIVE LOGITS
     initially
    1.20
     until
    1.08
     Initially
    1.08
    Initially
    1.08
     inicialmente
    1.06
    最初は
    1.04
    until
    1.03
    Until
    0.96
     awalnya
    0.95
     Until
    0.94
    Act Density 0.259%

    No Known Activations