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

    Neuronpedia

    INDEX
    Explanations

    edge cases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     opposites
    1.79
    ть
    1.74
     fissures
    1.73
     knots
    1.73
     codons
    1.72
     Aloha
    1.69
    ته
    1.68
     lumens
    1.68
     bouquets
    1.66
     germs
    1.66
    POSITIVE LOGITS
    ت
    2.22
    al
    2.19
    EN
    2.19
    т
    2.19
    ao
    2.14
    iq
    2.13
    essere
    2.11
    elijk
    2.08
    ੀ
    2.08
    ud
    2.06
    Act Density 0.067%

    No Known Activations