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

    Neuronpedia

    INDEX
    Explanations

    digits and building blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     turned
    0.48
     had
    0.48
     copyright
    0.45
     ul
    0.43
     ah
    0.43
     disen
    0.43
    tit
    0.43
     ambulance
    0.43
     d
    0.42
     ap
    0.42
    POSITIVE LOGITS
    싱
    0.43
    後ろ
    0.43
    紗
    0.40
    哃
    0.40
    ானி
    0.40
    に向
    0.39
     Специа
    0.39
    に関連
    0.39
     للح
    0.38
    実感
    0.38
    Act Density 0.001%

    No Known Activations