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

    Neuronpedia

    INDEX
    Explanations

    knowing and understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ج
    1.11
    кту
    0.98
     \
    0.96
    0.96
     $$\
    0.88
    بيه
    0.87
    ني
    0.84
     antennes
    0.84
    âni
    0.84
     In
    0.83
    POSITIVE LOGITS
    の
    1.21
    RO
    1.14
    1
    1.13
     the
    1.11
    า
    1.07
     and
    1.03
    ,
    1.02
    의
    0.98
    ↵
    0.98
    3
    0.97
    Act Density 0.127%

    No Known Activations