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

    Neuronpedia

    INDEX
    Explanations

    detecting negation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     without
    0.56
     Without
    0.46
    Without
    0.44
     tanpa
    0.42
     only
    0.42
     Just
    0.41
     Only
    0.41
     WITHOUT
    0.41
     seulement
    0.41
    without
    0.40
    POSITIVE LOGITS
     hề
    0.49
     gefähr
    0.38
    )*
    0.37
     capsid
    0.37
    ḫ
    0.36
     Berkshire
    0.36
     стр
    0.36
     खतरे
    0.35
    ropathy
    0.35
    ️⃣
    0.35
    Act Density 0.098%

    No Known Activations