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

    Neuronpedia

    INDEX
    Explanations

    before making

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     guesswork
    0.90
     mistakes
    0.89
     grammatical
    0.86
     선택
    0.83
     errores
    0.83
    ቤ
    0.81
     guesses
    0.80
     unsatisfactory
    0.80
     cuál
    0.79
     violations
    0.79
    POSITIVE LOGITS
     लेते
    0.80
    Q
    0.78
     any
    0.77
    nas
    0.75
    ana
    0.71
     anytime
    0.71
    Working
    0.70
    change
    0.70
    ε
    0.70
    Yeni
    0.70
    Act Density 0.003%

    No Known Activations