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

    Neuronpedia

    INDEX
    Explanations

    game winning strategies

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .float
    -0.09
     multiplic
    -0.08
     изд
    -0.08
     float
    -0.08
     мад
    -0.08
     Zit
    -0.07
    .lib
    -0.07
     jit
    -0.07
     zweimal
    -0.07
     δεύτε
    -0.07
    POSITIVE LOGITS
     analyzed
    0.10
    (pattern
    0.09
     paradigma
    0.09
    ıklı
    0.08
    Patterns
    0.08
     padrões
    0.08
     invariant
    0.08
     Patterns
    0.08
     판단
    0.08
    ýin
    0.08
    Act Density 0.016%

    No Known Activations