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

    Neuronpedia

    INDEX
    Explanations

    displacement or difference value

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     in
    1.20
    ill
    0.98
    ending
    0.97
    erty
    0.96
    umping
    0.95
    い
    0.94
    ido
    0.93
    ंद
    0.93
    ergic
    0.93
    eres
    0.92
    POSITIVE LOGITS
    is
    1.63
     Offset
    1.16
     Ciências
    1.16
    kách
    1.15
    to
    1.12
     Incluso
    1.12
    𝓂
    1.10
     grau
    1.09
     Jü
    1.07
     rápido
    1.06
    Act Density 0.003%

    No Known Activations