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

    Neuronpedia

    INDEX
    Explanations

    choose to learn patterns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Το
    0.92
     Το
    0.92
    То
    0.84
     tot
    0.84
    TOT
    0.82
    το
    0.81
    ofen
    0.79
     Tổ
    0.79
    tox
    0.78
    ΤΟ
    0.76
    POSITIVE LOGITS
     να
    1.20
     anzu
    1.05
     để
    1.01
     להת
    0.98
    ที่จะ
    0.94
     щоб
    0.94
     să
    0.91
     чтобы
    0.89
     einz
    0.83
     outlandish
    0.83
    Act Density 0.081%

    No Known Activations