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

    Neuronpedia

    INDEX
    Explanations

    use title

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     leggera
    0.46
     Tijdens
    0.44
     resumo
    0.42
     Quelles
    0.39
     নেতাকর্মীরা
    0.39
    ূর্ত
    0.39
     profil
    0.39
    }-\
    0.39
     ansatz
    0.38
     resumen
    0.38
    POSITIVE LOGITS
     addressed
    0.86
    addressed
    0.72
     recipient
    0.67
     addresses
    0.64
     PO
    0.62
    Dear
    0.60
     Addressing
    0.60
     addressing
    0.60
     Dear
    0.59
    attention
    0.59
    Act Density 0.033%

    No Known Activations