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

    Neuronpedia

    INDEX
    Explanations

    speaking verb ending quotes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    々は
    0.87
     valores
    0.85
    または
    0.85
    もしくは
    0.82
    ма
    0.81
     importants
    0.81
     coriander
    0.81
     relevante
    0.80
    又は
    0.80
    ヱ
    0.80
    POSITIVE LOGITS
    it
    0.98
    ner
    0.89
    lia
    0.85
    driven
    0.82
    ’.
    0.80
    at
    0.78
    crime
    0.75
     racconta
    0.75
    loan
    0.75
    partisan
    0.75
    Act Density 0.000%

    No Known Activations