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

    Neuronpedia

    INDEX
    Explanations

    He followed by avier or heave

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    形の
    0.41
    форми
    0.41
     ಕಾನೂ
    0.40
    igenza
    0.39
    ಿಸಿದ
    0.39
    셈
    0.38
    落ち
    0.38
     tambahan
    0.38
    fice
    0.38
    岚
    0.38
    POSITIVE LOGITS
     ヘ
    0.64
     хе
    0.62
     he
    0.61
     Heg
    0.58
    Hem
    0.58
     hems
    0.55
     HE
    0.55
     heap
    0.55
     heg
    0.53
    He
    0.53
    Act Density 0.027%

    No Known Activations