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

    Neuronpedia

    INDEX
    Explanations

    code and programming terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    fandom
    -0.79
     stock
    -0.78
     goles
    -0.77
    рат
    -0.77
    ulier
    -0.77
     tribus
    -0.77
     prisa
    -0.77
     CRIST
    -0.76
    ☆☆☆
    -0.75
     erit
    -0.74
    POSITIVE LOGITS
    instantiate
    1.01
     instantiate
    0.90
    Instantiate
    0.75
     destroy
    0.70
     POSITION
    0.68
     participated
    0.66
     Laid
    0.65
    POSITION
    0.65
     intercom
    0.64
     trận
    0.63
    Act Density 0.006%

    No Known Activations