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

    Neuronpedia

    INDEX
    Explanations

    rumors, speculations, and expert opinions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Implemented
    0.71
     valuable
    0.67
    पया
    0.66
    ще
    0.65
    处理
    0.65
    Detectable
    0.65
     unknowingly
    0.64
     நம
    0.64
    判定
    0.63
     unconsciously
    0.63
    POSITIVE LOGITS
     allegations
    1.09
     pundits
    1.03
     rumors
    1.01
     commentators
    0.94
     разгово
    0.93
     rumours
    0.90
     allegation
    0.87
     अटक
    0.87
     speculate
    0.86
     accusations
    0.85
    Act Density 0.777%

    No Known Activations