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

    Neuronpedia

    INDEX
    Explanations

    effective content description

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     referring
    0.61
     manifesting
    0.60
     iff
    0.59
    aretro
    0.59
     angezeigt
    0.58
    levance
    0.58
     implying
    0.57
     rifer
    0.55
    indicated
    0.55
     refinancing
    0.54
    POSITIVE LOGITS
     accurately
    1.09
     effectively
    1.04
     successfully
    1.03
     comprehensively
    1.03
     attempt
    1.01
     thoroughly
    0.93
     efficacement
    0.91
     eloquently
    0.91
     admirably
    0.91
     эффективно
    0.91
    Act Density 0.188%

    No Known Activations