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

    Neuronpedia

    INDEX
    Explanations

    I've, I'd, I'm, I do

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ≠
    0.82
     AGAIN
    0.76
     ==
    0.76
     matem
    0.73
     repet
    0.70
     somit
    0.69
    //#
    0.68
     ===
    0.67
    そもそも
    0.66
     igjen
    0.65
    POSITIVE LOGITS
     আমিও
    1.02
     wouldn
    1.00
     wouldnt
    0.97
    經驗
    0.92
    建議
    0.90
     advise
    0.89
     struggled
    0.89
     advice
    0.88
     পরামর্শ
    0.86
    expérience
    0.85
    Act Density 0.037%

    No Known Activations