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

    Neuronpedia

    INDEX
    Explanations

    discussion topics and contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     equals
    0.39
    ousine
    0.38
    educt
    0.38
    WinCounter
    0.37
    ullin
    0.37
    aints
    0.37
    oston
    0.36
    Products
    0.36
     त्र
    0.36
     eşit
    0.35
    POSITIVE LOGITS
     normative
    0.40
     utilisateur
    0.39
    ↗
    0.38
    옌
    0.38
    復
    0.38
     usuário
    0.37
     quantized
    0.37
     istot
    0.37
     workflows
    0.37
    コンビニ
    0.37
    Act Density 0.000%

    No Known Activations