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

    Neuronpedia

    INDEX
    Explanations

    electric, enemies, survive, make

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     itaque
    0.34
     vě
    0.33
     revenir
    0.32
     beware
    0.31
     silence
    0.31
     živ
    0.31
     awake
    0.31
     Medicine
    0.31
     divisor
    0.30
     ampl
    0.30
    POSITIVE LOGITS
    ｩ
    0.40
    <0x93>
    0.40
    âu
    0.39
     শ্
    0.39
    dDays
    0.38
    irma
    0.38
     Darth
    0.37
    etze
    0.37
    cluding
    0.36
    きましたが
    0.36
    Act Density 0.055%

    No Known Activations