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

    Neuronpedia

    INDEX
    Explanations

    introduces comparisons or explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    解释
    0.31
    Let
    0.31
     tens
    0.30
    Firstly
    0.30
    southern
    0.30
    composite
    0.29
     dijel
    0.29
     Vamos
    0.29
    combined
    0.29
     우선
    0.29
    POSITIVE LOGITS
     comparisons
    0.57
     Comparison
    0.50
     comparison
    0.49
     comparaison
    0.49
     COMPAR
    0.48
     comparación
    0.47
    Comparison
    0.46
    COMPAR
    0.46
     Comparisons
    0.45
     срав
    0.44
    Act Density 0.002%

    No Known Activations