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

    Neuronpedia

    INDEX
    Explanations

    personally identifiable information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    эффици
    0.89
     unidirectional
    0.89
    },\\
    0.87
     sít
    0.86
     vestibular
    0.84
     года
    0.84
    псо
    0.84
    ्रिया
    0.83
    циями
    0.81
     Kähler
    0.81
    POSITIVE LOGITS
    ب
    0.99
    e
    0.80
     dessin
    0.80
     faire
    0.79
    four
    0.75
    實際
    0.75
     deux
    0.75
     sauf
    0.75
     locaux
    0.74
     de
    0.74
    Act Density 0.000%

    No Known Activations