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

    Neuronpedia

    INDEX
    Explanations

    amounts and ranges

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Beyond
    0.44
     even
    0.42
    Even
    0.42
     similarly
    0.41
     beyond
    0.39
     đủ
    0.39
     likewise
    0.38
    agir
    0.37
    น
    0.37
     demais
    0.37
    POSITIVE LOGITS
     гораздо
    0.65
     훨씬
    0.51
    closer
    0.43
    훨
    0.42
    раздо
    0.42
     nearer
    0.41
     намного
    0.41
    кает
    0.40
     hingegen
    0.40
    лась
    0.40
    Act Density 0.062%

    No Known Activations