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

    Neuronpedia

    INDEX
    Explanations

    averaged out

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     perils
    0.50
     answers
    0.49
     coins
    0.49
     電気
    0.48
    coins
    0.48
    astos
    0.47
     エネルギー
    0.47
     frizz
    0.47
     silverware
    0.46
     songs
    0.46
    POSITIVE LOGITS
     беше
    0.46
    ARDO
    0.46
    A
    0.46
    elit
    0.44
    Ay
    0.44
    É
    0.43
    El
    0.43
    Alg
    0.43
    Version
    0.43
    Ej
    0.43
    Act Density 0.000%

    No Known Activations