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

    Neuronpedia

    INDEX
    Explanations

    will make negative outcome

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     somewhat
    0.35
    稍微
    0.34
     এসেছে
    0.32
     Some
    0.32
     helper
    0.32
     തന്നെയാണ്
    0.31
     dùng
    0.30
    を用いて
    0.30
    封装
    0.30
     slightly
    0.30
    POSITIVE LOGITS
     disastrous
    0.69
     irrepar
    0.63
     jeopardize
    0.57
     irreparable
    0.57
     injustice
    0.55
     worse
    0.54
     endangering
    0.52
     needlessly
    0.52
     jeopard
    0.51
     irrevoc
    0.51
    Act Density 0.764%

    No Known Activations