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

    Neuronpedia

    INDEX
    Explanations

    if clauses and hypotheticals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    所谓
    0.46
     vér
    0.43
     выяс
    0.41
     contém
    0.41
    してます
    0.41
    是否
    0.40
     evidentemente
    0.40
    ಡಿ
    0.40
     liệu
    0.40
    อื่น
    0.40
    POSITIVE LOGITS
     hypot
    0.70
     Hypot
    0.59
     were
    0.58
     had
    0.57
     Were
    0.55
     asked
    0.54
     hypothetical
    0.52
     Could
    0.52
    had
    0.51
     unlimited
    0.51
    Act Density 0.014%

    No Known Activations