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

    Neuronpedia

    INDEX
    Explanations

    how something works

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    писание
    1.01
    ளில்
    0.99
    centaje
    0.98
    یاء
    0.97
    ulent
    0.95
    glicherweise
    0.92
    आइए
    0.92
    compl
    0.91
     considere
    0.91
    aneous
    0.90
    POSITIVE LOGITS
    ки
    1.47
    ्डा
    1.38
    evil
    1.37
     bunch
    1.36
    로
    1.36
     exacting
    1.35
    ва
    1.35
     egress
    1.34
     lähe
    1.34
    es
    1.34
    Act Density 0.000%

    No Known Activations