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

    Neuronpedia

    INDEX
    Explanations

    speculation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ра
    1.10
    ，
    1.02
     Само
    0.95
    ،
    0.93
    но
    0.90
     Пере
    0.90
    تا
    0.88
    ول
    0.87
    ных
    0.87
    рати
    0.87
    POSITIVE LOGITS
     speculate
    1.36
     on
    1.32
    t
    1.16
     speculations
    1.15
     speculation
    1.08
    ENT
    1.02
     speculative
    1.02
     speculating
    1.02
    \
    0.99
    AN
    0.98
    Act Density 0.003%

    No Known Activations