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

    Neuronpedia

    INDEX
    Explanations

    explaining why

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     отличается
    1.09
     differs
    1.08
     отличаются
    0.91
     restricts
    0.85
     altru
    0.84
     differed
    0.84
     inhibits
    0.82
     удалить
    0.80
     tissue
    0.79
     পদার্থ
    0.79
    POSITIVE LOGITS
    ாடி
    1.11
    se
    1.04
    ა
    1.03
     Recordemos
    1.02
    lında
    0.99
    说
    0.99
    ただ
    0.98
    ES
    0.97
    lah
    0.96
    los
    0.94
    Act Density 0.007%

    No Known Activations