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

    Neuronpedia

    INDEX
    Explanations

    nicknames and alternative names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     notwend
    0.47
     lehető
    0.41
    があれば
    0.41
    叕
    0.41
     toekomst
    0.40
    朤
    0.40
    mW
    0.40
     cuidadosamente
    0.39
     apoyar
    0.39
     entrop
    0.39
    POSITIVE LOGITS
    日本では
    0.50
     colloqu
    0.45
     when
    0.43
     оши
    0.41
     colloquial
    0.41
     informally
    0.39
     Collo
    0.39
    anners
    0.39
     referring
    0.39
    refer
    0.39
    Act Density 0.131%

    No Known Activations